JPH05507210A - Multi-quantum NMR using frequency modulated chirped pulses - Google Patents

Abstract

(57) [Summary] This bulletin contains application data before electronic filing, so abstract data is not recorded.

Description

[Detailed description of the invention] Multi-quantum NMR using frequency modulated chirb pulses The present invention applies a first π/2 pulse to a nuclear spin system to which a magnetic field is applied; The second pulse occurs after the defocusing time 1[1τ following two pulses has elapsed. occurs after one π-pulse and another time interval τ following the first π-pulse. an RF pattern comprising an elementary sandwich consisting of a second π/2 pulse In NMR pulse experiments by irradiating the first sequence of Excite two-quantum coherence and transfer the excited multi-quantum coherence to the nuclear speed. By irradiating the magnet system with a second series of RF pulses, it reconverts to longitudinal magnetization and becomes free. This invention relates to a method of inducing induced damping.

Methods of the type mentioned above are described by D. B.E.

1'einkamp) Adv, Mega, Re5oa, 11.11 1 (1983) scientific publication.

Many modern NMR methods rely on transverse magnetization, or more generally pIk coherence. A coherent chain in which a spin (or set of spins) moves from one spin (or set of spins) to another. It takes advantage of the phenomenon of space movement. Some of the best known examples include correlation spectroscopy and There is quantum NMR. Many of these experiments were performed using the density operator (tJensIIyop In order to guide the rotation of the e-razor, the so-called "bard" based on the widespread use of In reality, such pulses are truly offset. is rigid enough to be offset-independenl. There is no off-resonance effect (tilted position). (effective magnetic field) must be taken into account. As in correlation spectroscopy (CO3Y) In the simplest experiments, off-resonance effects are simply a matter of appropriate data processing. It only results in a relatively harmless phase error that can be easily corrected.

However, in more sophisticated experiments, the error in successive pulses is It often has a cumulative effect that can lead to complete failure of the healing move process. Hidden Therefore, in multi-quantum NMR according to the above quotation, coherence consists of three monochromatic Excited by a sandwich of line frequency pulses (π/2−τ−π−τ−π/2) ing. If the effective magnetic field is tilted, this sandwich can fail completely. Ru. This point is supported by Levitt et al., who proposed another series incorporating composite pulses. and Ernst (LetNl and Etnsf), Mo1. Ph ys, 50.1109 (1983). child These series show that several consecutive rotations have better overall performance than a single rotation. It is based on the idea that it can be designed to have a certain amount of motion. composite pulse Constraints apply to the frequency window within which they are restricted. typically over an offset range comparable to the RF amplitude. The point is to show off your sexuality.

In the following, from the initial state where the magnetization is aligned with the magnetic field, the magnetization changes with respect to the magnetic field. The pulse that converts to a new perpendicular state is called a ``π/2 pulse,'' whereas the transverse magnetic A pulse that converts magnetization into transverse magnetization in such a way as to produce spin echoes is called a π pulse. ” I will call it.

Two quantum components of a pair of weakly coupled spin 1/2 nuclei, such as Levitt and Ernst Let's concentrate on the light method. From the point of view of spectroscopic recognition of molecular structure, this type of One of the most promising experiments possible is J. Am. Chem. So. Apax et al. (A, Ba!) in c, 102.4849 (1980). It is known as INADEQUATE from e1at, ). to this method This allows us to determine the entire carbon skeleton of an organic molecule simply by examining the signal in a two-dimensional spectrum. It becomes possible to identify. This method uses two quantum co-ordinates over an arbitrary spectral width. If there were no difficulties in obtaining reliable excitation of heerence, a larger person There is no doubt that he will be impressed.

Therefore, a primary objective of the present invention is to use relatively small amplitude RF pulses to As mentioned at the beginning, we can efficiently excite multi-quantum coherence over a wide offset range. The purpose is to provide different methods.

According to one aspect of the invention, this purpose is to the pulses are chirped pulses each having a duration t, during which , the frequency of the excitation RF field varies from the lower limit frequency ω□, 11 to the upper limit frequency ω1.1. Nimata On the contrary, is a chirped pulse that is swept in a monotonous relationship with respect to time, and the basic the duration t, /2 and 2 to 4 times the amplitude of the first π/2 pulse. A fourth chirb pulse with an amplitude of is achieved by the following method.

Multi-quantum coherence is the fundamental sandwich of three RF pulses for spin systems. However, the subsequent fourth chirb pulse is generated by the multi-quantum coherence the phase of the signal so that the phase is nearly offset-independent over a large bandwidth. reimage the sea urchin. The sequence of chirped pulses has large offsets and tilts. It can be considered as an alternative to composite pulses that overcome the effects of magnetic field effects. can. Book J! The conditions for reimaging the magnetization vector are set by the bright chirped pulse sequence. i.e. a very large bandwidth that would otherwise account for the phase distortion can be kept negligible over It combines the advantages of chromatographic techniques with those of CW spectroscopy. Moreover, it is a solid color. By using a chirb pulse instead of a bright pulse, multi-quantum coherence can be improved. The pulse amplitude required for excitation is considerably reduced, which is significantly reduced in vivo ( In the field of applications (in-vivo), for example, NMR imaging or It has particular advantages in MR-in-vivo spectroscopy. do.

In a preferred embodiment of the method of the invention, an elementary sandwich of three RF pulses immediately followed by the fourth reimaging chirped pulse, thus maximizing the relaxation time. This means that the loss of multi-quantum coherence due to relaxation is smaller. , thus meaning that the sensitivity of the method is greater.

The method of the invention is less sensitive to the amplitude of said first chirped π pulse. However, in most practical cases, the amplitude of the first π pulse is reduced to the amplitude of the first π/2 pulse. coverage by selecting an amplitude of 2 to 4 times, preferably about 2.8 times, the amplitude of the It will be done.

The amplitude of the second π/2 pulse in the basic sandwich is equal to that of the first π/2 pulse. The amplitude of the fourth chirped pulse should be approximately the same as the amplitude of the fundamental sandwich pulse. should be selected to be approximately 3.3 times the amplitude of the first π/2 pulse in the first inch.

Another object of the present invention is to solve the problem of unobservable multi-quantum coherence (e.g., 2 Retrieve the observable single-quantum coherence (of spins A and M in a spin system) That's true. According to another aspect of the invention, this means that the first The phase depvising time following the multi-quantum echoes generated by the sequence of A second series of RF pulses is started after an interval of 1, and the second series of RF pulses is , the frequency of the RF field changes from the lower limit frequency ωIFMIN to the upper limit frequency ω1, □ in time. The chirped reimaging π-pulse with duration t,/2 is swept in a monotone relation to and a subsequent chirped π/2 pulse of duration t. In a preferred embodiment of the method, the second series of RF pulses further includes chirping. The duration τ,=t,/2 applied after the delay time τ2 following the π/2 pulse is also another chirped π pulse, and the resulting spins 221 are also It is detected after 3 at the final delay time following another chirped π pulse.

This produces a pure phase spectrum, i.e. pure absorption or pure dispersion, in the ω2 dimension. This takes into account the possibility of collecting the waste at a later date.

In the case of excitation of two-quantum coherence between spin A and spin M, the delay time τ2 is τ2zO, and the final delay time τ3 will be τ3””t*. You can choose to This creates an antiphase structure in the two-dimensional spectrum. A doublet is obtained. If the delay time τ2 is 2 = (2J AM) -1 Then, the final delay time τ3 becomes τ2+τ, +τ3χJAkl-' is chosen so that J AIJ is the scalar resultant of the coupling between spin A and spin M. If we represent the sum constant, the amplitude of the signal as a function of the offset will be smoother. , ie the undesired frequency dependence of the signal becomes smaller.

In order to obtain a doublet with an in-phase structure in a two-dimensional spectrum, the time so that the delay 2 is τ2χ(4JAM)-1 and the final delay τ3 is τ2+ τ, +τ33(2JA,)-1.

Recovering observable single-quantum coherence from multi-quantum coherence is a It can also be achieved by another embodiment of the method according to the present invention, but in this embodiment, , the three RF pulses of the elementary sand inch each have a duration t, pulse, during which the frequency of the excitation RF field is at a lower limit frequency ω11. ．． to upper limit frequency ω01. swept up to f or vice versa in a monotonic relationship with respect to time. and, after the development time tl, the basic sand inch is exactly time-reversed sandwiches, i.e. each having duration t, and each other After a time interval, the upper limit frequency ω□3.8 to the lower limit frequency ωlFm. to Three chirve pulses of type π/2, π and again π/2 swept or vice versa followed by

Observation of Ig generated by propagalor V The final transition to possible one-quantum coherence is due to the chirped π/2 pulse and its followed by half the duration of the preceding chirb π/2 pulse and 2 to 4 times, preferably is achieved by an echo sequence consisting of chirped π pulses with 2.8 times the amplitude. Good too.

Furthermore, the amplitude of the first π pulse in the basic sandwich inch is the same as that of the first π/2 pulse. should be selected to be 2 to 4 times the amplitude of, preferably about 2.8 times, The amplitude of the second π/2 pulse in the basic sandwich is the amplitude of the first π/2 pulse. It should be approximately the same as the width.

By selecting a very short expansion time t1χO, one-dimensional r A spectrum is obtained.

Two-quantum coherence (DQC) and zero-quantum coherence excited in the fundamental sandwich In order to reimage (telocas) the lens (ZQC), the finite expansion time It is possible to apply a hard reimaging π pulse at the beginning or end of the interval t1. .

The reimaging of multi-quantum coherence also occurs after the fundamental sand inch during the unfolding time t1. duration t, /2 and 2 to 4 times the amplitude of the first π/2 pulse of the fundamental sandwich inch. is achievable when followed by a fourth chirped pulse with an amplitude of . most actual In the case of The wave pulse is selected such that its amplitude is equal to the first π/in the fundamental sandwich inch. The amplitude is selected to be approximately 3.3 times the amplitude of the two pulses.

In a preferred embodiment of the method according to the present invention, multi-quantum coherence can be observed. The second sequence of RF pulses to reconvert to one quantum coherence is the chirp π /2 pulse followed by half the duration of the preceding chirp π/2 pulse and It consists of a chirped π pulse with an amplitude of 2 to 4 times, preferably about 2.8 times.

The acquisition of the free induction decay signal following the second series of RF pulses is performed at the onset of this pulse series. Immediately or, in some cases, preferably, at the bottom of the spin-echo signal. It may start at large intensities, and in the latter case, by Fourier transform of the echo signal The resulting spectrum is essentially free from phase dispersion.

In the normal case, the frequency sweep of the RF excitation radiation is already carried out at a constant sweep rate. is linear in time. However, in special cases, the limiting frequencies ω0, 5 and It may be helpful to perform a nonlinear but monotonic frequency sweep between ω□−1. stomach.

The present invention is applicable to various NMR methods such as two-dimensional transformation spectroscopy, homonuclear and heterogeneous uncorrelated spectroscopy. Since it is useful for pulse spectroscopy applications, it has the above-mentioned spectroscopic capabilities and according to the present invention. An NMR pulse spectrometer with a device that allows the operating mode is also considered a subject of the invention. It will be done.

Further details, aspects and advantages of the invention will emerge from the following description based on the drawings. However, in the drawing, Fig. 1 shows the time dependence of the RF frequency (top), the instantaneous Figure 2 shows the time RF phase (middle) and the schematic nuclear magnetic resonance spectrum (bottom); , a) Basic sand-in of three chirped pulses designed for broadband multiquantum excitation RF vibration γB1 and RF frequency ω□ for Q, b) Inverse sum term 121A□I’XYl C) shows the time dependence of the magnitude of the two quantum term IDQC+; Fig. 3 2 as a function of the initial phase φ3 of the third pulse during the fundamental sandwich inch. Two-quantum coherence (solid line) and zero quantity excited by the fundamental sand inch of The magnitude of the child coherence (dashed line) is shown in the figure, a) All parameters are as shown in Figure 2. and b) the effective flip angle β3 of the third pulse is reduced to π/4. It is: Figure 4 shows spin A (solid line) and spin M (dashed line) as a function of phase φ3. The contribution to a) two-quantum coherence and b) zero-quantum coherence of The free-ruff angle of Noculus in 3 is β = π/2; Figure 5 is a graph of the contribution to two-quantum coherence arising from spins A and M. the law of nature: Figure 6 shows a) two-quantum coherence after the basic three-pulse shift of Figure 2. and b) the magnitude of zero quantum coherence l ZQCl: Figure 7 shows a calculation based on a) numerical simulation and b) pulse/deployment model. The two-quantum core excited by the elementary sandwich with β3=π/2 by calculation Indicates the magnitude of Heerence IDQC+; Figure 8 shows the excitation followed by the Chirb reimaging noculus acting on the two-quantum coherence. RF amplitude γB+, (top) and RF frequency ω, , (bottom) for the sand-inch case Figure 9 shows the time dependence of the two-spin quantity as a function of the relative offset of the two spins. Figure 2 shows a) magnitude IDQcI and b) phase Φ of the child coherence. The series shown in Figure 8 is indicated by a solid line, and the series shown in Figure 8 is indicated by a broken line; Figure 10 shows a) the reconversion sequence of the chirped pulse and b) the corresponding Figure 11 shows the offset sequence. arises from the reconversion from two-quantum coherence as a function of Ω6 and Ω2. Antiphase-quantum term a) Size 12 Axy Axy Mach 1 and 121% I”xyl and b) Indicates the phase φ and φ. Figure 12 shows two-quantum coherence using time reversal. RF for excitation and reconversion of Figure 13 shows the time dependence of the pulse sequence; a) R similar to that in Figure 12; RF frequency ω of the F pulse sequence and the corresponding b) antiphase-quantum coherence 2I A□I　xv and C) Show the time dependence of the magnitude of the two quantum term DQC; Figure 14 shows the reconverted zeolite from two-quantum coherence as a function of offset. - Mann ring IA□ and 1%; and Figure 15 shows the bond J　A&l=5　5 Hz and JA-&1- = 15 Hz, two two-spin subsystems (2- Protons with enlarged regions showing spin 5 ubsys+ems) This is an experimental two-quantum spectrum of a pin system.

Consider a scalar coupled two-spin system under a frequency modulated pulse. transmitter In a rotating coordinate system synchronized with the instantaneous frequency of nian): H(t)=Ho(t)+HR.

= Ω (t) I + Ω (t) I2 + 2 JI -1^ ZM 10γEl, C(IQ+I;’)eosφ+(I’,+I’,)srnφko, (1) where ΩA(1)=ω. 1-ω, (1) is the time-dependent transmitter frequency is the offset of spin A with respect to wave number. In this coordinate system, the RF phase ψ is time-independent, but the coherence movement path is selected by phase cycling. can be shifted from one pulse to the next to select. RF radiation γB.

is assumed to be constant during the frequency sweep. Hamiltonian iaI+) is assumed to be open time independent over short time intervals, Ltou vtlle-won Neuman's formula 6 formula % (2) is integrated over the interval [1/, 1/+τ], and a(t’+t”)=exp (-iH(t')t)cr(t')exp(jH(t')r) (3) can be obtained.

In this way, the time expansion of the density operator σ becomes an instantaneous Hamiltonian (Hamil+ onian) H(t') can be calculated by stepwise transformation.

H(t’) includes non-commutative contributions from RF radiation, chemical shifts and scalar bonds. Therefore, each step is a numerical diBonalization ion). Alternatively, the Conitari transformation of equation (3) is the higher option of exp(-iHot)exp(-iHipr)(4i)+τ [Address section] Therefore, for different parts of the Hamiltonian (Hi+n1ljonian) can be decomposed into successive rotations.

For small τ values, terms with quadratic and higher orders of τ can be ignored.

e x p (1 (HO + HIIF) τ) ze x p (-i Hot) e x l − i Harτ) (4b) Thus, for each interval, HO and The effects of H and Hl may be calculated continuously. Furthermore, by introducing the weak coupling assumption, Eq. The scalar combination order in (1) may be reduced by 2πJI-I8 towns. This results in We can expand the density operator in the product of angular momentum operators.

We perform simulations based on both the Cartesian product operator method and the explicit matrix notation of the density operator. ration was carried out. Typically about 2000 intervals of 20k in 16ms It is necessary to describe the effect of a pulse covering a sweep range of Hz.

As an alternative to the accelerated rotational coordinate system used in equation (1), a constant reference frequency ωp A normal coordinate system rotating at a+ may also be considered. In this coordinate system, φ□( +)=Jω1. +(+)dt, so the linear frequency sweep is It means that the phase changes quadratically as a function of time. In this coordinate system is the first of the transmitters at the beginning of the sweep to select the coherence moving path. It is possible to shift the phase φ, (0).

The conventional rotating coordinate system is a natural reference frame for the approximate description of the chirb pulse, Helps guide intuition. This description means that when the spin passes through resonance exactly It is assumed that the sensor is affected by the RF field. As a result we are, in effect, pure It has instantaneous semi-selective pulses separated by a typical development interval. for simplicity This picture will be called the 1-pulse/expansion model.

Since the J-coupling is always very small compared to the RF amplitude γB1, both components are affected simultaneously, hence the use of the expression “semi-selective pulse”. It can be assumed that This is a combination of Feretch (Fer+et1) and Ernst (Er+et1). must be distinguished from the methods advocated by In describing a rapid passage, only one transition is affected at a time. Using a cascade of truly selective pulses to give

FIG. 1 shows the relationship between frequency and phase. For a linear frequency sweep, the spin A The phase φ6 of the semi-selective pulse used is secondarily determined by the passage during resonance as shown in FIG. At the instant tA, φ = TbtA'', or equivalently, at the offset, φ = ΩA'/ b. Associated. Sweep speed (sweep rjre) b is the sweep range and b=ΔIJJ,/l, expressed in rad seconds″2. Ru.

The resonance of spins A and B of the chirb pulse with a linear frequency sweep is The wave number (top) coincides with the chemical shifts Ω and Ω8, and the instantaneous RF phase (middle) coincides with φ and φ3 are passed at times tA and I5.

In a system containing an isolated spin I = 1/2, the pulse/expansion model has two It is quite sufficient to predict the formation of an echo due to a sequence of chirped pulses. deer However, in coupled spin systems, this model only gives a qualitative picture. the below described In this paper, the analysis results obtained from the pulse/expansion description are transformed into a Hamiltonian (Ham stepwise diagonalization of i-th onian) or operator type transformation (oper[.

The exact system obtained by Let's compare it with simulation.

The basic sandwich consisting of three chirb pulses in Figure 2a is a multi-quantum coherent The (π/2−τ−π−τ−π/2) sandin is commonly used to generate It is built on the same principle as Chi. All three chirb pulses had the same duration Therefore, they must have the same sweep race. the Amplitude adjusted to achieve analog of either π/2 pulse or π pulse It must be. The Chirb π/2 pulse has a relatively weak RF amplitude. characterized, so an adiabatic condition was established in which the magnetization was locked to the effective magnetic field through the sweep It is intentionally violated so that it does not remain as it is. Chirped π pulses, on the other hand, RF amplitude sufficient to allow thermal reversal or, as in this case, reimaging of the transverse magnetization. have. In adiabatic reversal, the magnetization remains locked along the effective magnetic field. On the other hand, due to reimaging, the sweep spacing remains perpendicular to the effective magnetic field. Second In figure a, the first chirped π/2 pulse generates transverse magnetization starting from thermal equilibrium. the second pulse reimages the effect of the offset (chemical shift), but the same That of the nuclide scalar coupling does not reimage, and the third pulse is an antiphase-quantum magnetic Converts quantum coherence into multi-quantum coherence. Relative phase is two-quantum or zero-quantum coherence must be optimized for maximum creation of lenth. traditional sandwich inch The interval between the beginnings of the first and third pulses is (2JA,)-' as in must be ordered.

Figures 2b and 20 show antiphase-quantum terms and two-quantum coherence in a two-spin system. This shows the time dependence of nu. The phase behavior is quite complex as it refers to the accelerated rotational coordinate system. , so we have the expansion ( (see definition in Table 1). The antiphase term is approximately 5 in (πJ AIJ increases spontaneously under the influence of scalar coupling with an envelope proportional to t).

During the third pulse of the sand inch, the antiphase term partially transforms into two-quantum coherence. will be replaced. Figure 2 shows how l　DQC1 term is initially 12I^IM, vl , and then from 12xAxvx”tl, two successive stages It shows whether the amount increases with each floor. Whether positive interference of these two contributions can be obtained is determined by the relative phases of the three pulses. Phase of the second pulse φ2 The shift of 2φ2 of one quantum coherence in the reimaging period after the second pulse only induces a phase shift at the end of the chain, so that the third pulse of the sandwich It is possible to show that it is sufficient to consider the effect of the phase φ3 of the Ru.

The RF frequency is the chemical shift ΩA/2π=7kHz at the time indicated by the arrow. 2 and ΩM/2yt=13kHz.

Simulation parameters 'J AIJ = 8Hz, t, = 16mm seconds, τ = 15 ms (therefore 2t, +2τ = 1/2JA,), sweep range 2 0 kHz, respectively effective flip angle π/2. three with π and π/2 γBz for continuous pulse = 309Hz, 860H! and 309H! It is. 3 The initial phases of the two pulses are φ1=φ2=0 and φ3=120°, the latter being the most selected for suitable excitation. Almost the maximum of two quantum terms in two successive steps Note the increase to large amplitude. Figure 3 shows the three pulse RF amplitudes as shown above. If selected, the optimal value of the two-quantum excitation is the phase φ3χ120” (±180’) Indicates that it corresponds to In fact, the phase requirements and the amplitudes of the π and π/2 pulses are interdependent. Various conditions must be examined for sufficient excitation of multi-quantum coherence. I can put it out.

In Figure 3a, all parameters are the same as in Figure 2, but in Figure 3b. In this case, the effective flip angle β3 of the third pulse is reduced to π/4. β3; For π/2, zero quantum excitation is very inefficient regardless of φ3. Be careful.

Figure 4 shows how the two contributions originating from spins A and M are combined with the third contribution. Add or cancel each other depending on the pulse phase ψ3 and effective flip angle β3 Show that. If you have a π/2 pulse, the contribution to zero quantum coherence is , always cancel each other out as shown in FIG. 3a. This means that the zero quantum coherence is π/ Consists of hard pulses that cannot be generated from antiphase magnetization by two pulses (π /2-τ-π-τ-π/2) series. However, if the RF By decreasing the width, the effective flip angle of the third pulse becomes β3 = π/4, and the RF phase is near φ3=0 or 180°, equal amounts of two-quantum and zero-quantum coherence lens (see figure 3b) can be obtained.

This behavior is again similar to classical experiments. Simple pulse/expansion discussed above Depending on the model, the behavior shown in Figure 3 cannot be predicted. Because it is 3 This implies that maximum two-quantum excitation occurs if all three pulses have the same phase. Because it is. Direct comparisons with exact simulations are difficult.

Figure 5 shows U, Eggenberger aI explaining the pulse/expansion model. ld G, Bodenhausen in Angew, Chell, + 02+ 392 (1990) and Anget, C’bem, In1 ．． The regulations discussed in the publication Ed, Engl., 29.374 (1990). It uses conventions to show "cartesian product operator expansion graph". . The two contributions to the spin and to the two-quantum coherence arising from M are separately shown. Zeeman order IA of spin A is first in phase - quantum coherence lA, , which evolves during the period 2τ+2t, +tAM and has an antiphase − The quantum coherence becomes 21’,,I”z, which finally becomes the two-quantum coherence. will be converted to On the other hand, I”z is transformed into IN, , which becomes the period 2τ+”jt between AM wholesalers and becomes 21A□1%, before being converted to two-quantum coherence. . Ideally, the two different periods 2τ+2t, ±tA&l shown in FIG. Both should be approximately equal to (2JA&1)-', which clearly indicates that tAM< It cannot be satisfied unless <(2JA,)-1.

In FIG. 5, the resonant frequencies are passed through at the points indicated by the vertical dotted lines. Each group The circles above and below the rough correspond to spins A and M, respectively. These intraday symbols Z + x / y and 1 are respectively operators 1. ．． 1. Or engineering 2. and unit performance Sanko (unit opera+.

r).

In addition to the two main paths shown in FIG. How much is enough to cause a transition from in-phase coherence to anti-phase coherence? If so, is there an additional small contribution to two-quantum coherence? Maybe.

These paths become even more important when tAM is comparable to (2JA,)-1. and in this case each chirped pulse taken in isolation has a significant amount of second It is possible to provide quantum coherence.

One-bond scalar coupling constants "Jcc" are usually all in the same range ( It is about 40H, but the chemical shift of the spin of carbon-13 (earbon-13) has a very large frequency range, typically 200 Ppm (600M 30 kHz in a Hz spectrometer. Depending on these circumstances Therefore, we designed a series of chirb pulses and evaluated its performance through simulation and experiment. It becomes possible to limit the guidelines for testing.

All three pulses in the excitation sandwich in Figure 2 have the same duration. , the magnitude of the multi-quantum coherence that occurs is off over most of the sweep range. It is set independent.

Figure 6 shows the offsets of zero and two quantum terms generated by a three-pulse excitation sequence. This indicates dependence on the target.

The effective flimbe angle of the third pulse is β3 = π/2 (actual m > or β3 = π/4 ( (dashed line). The offset Ω6 of spin A remains unchanged (ΩA /2π=5kHz), but Ω3 moves towards the right end of the sweep, i.e. Ω&l/2π varies from 5kHz to 20kHz. The chive pulse is the second As shown in the figure, it is swept from Ok Hz to 20 kHz in 16 milliseconds.

Figure 7 shows the exact calculations derived from the pulse/expansion model discussed above. It is compared with the analytical prediction. We used three different sweep speeds to Consider two-quantum coherence excitation according to the series of figures. Phase of increasing profile The tendency for the two intervals highlighted in Figure 5 to decrease with increasing offset is shown in Figure 5. This is because it is not possible to simultaneously optimize all pairs of spins. Chapter 7b Unlike the raw pulse/expansion description in the figure, the numerical simulation in Figure 7a , the so-called "edge effect" resulting from the finite range of the frequency sweep of the RF pulse eeffects) which accurately predicts J.

The offset Ω of spin A is fixed at ΩA/2π = 10 kHz However, ΩM/2π varies from 10 kHz to 40 kHz. chirp pal The signal is swept from OkHz to 40kHz with different sweep speeds. chirp The duration of the pulse, t, is shown. RF amplitude γB8 is adjusted proportionally to AIb It was done. The reduced performance for lower sweep speeds is due to the period t□ shown in FIG. This can be explained by the increase in . The edge effect (edge erfecf5) is , 1 depending on the pulse/deployment model! Note that it is not made clear.

The range of offsets over which the series is valid can be expanded by increasing the sweep speed. It may seem surprising that it is Noh. In fact, a fast sweep is , period tAu=lΩ, -Ω21/b for period 2τ+2t, (see Figure 5) It means being reduced. The shorter tAu, the shorter both periods 2τ +2t, ±tA□ can be close to (2JAM)-” regardless of the offset. This is closer to the ideal case. On the other hand, it tends to worsen with increasing sweep speed. Certain edge effects are important for sequences using fast sweeps. It does not constitute a fault that is too serious.

In the case of a simple adiabalic inversion, the edge effect The effect is to smooth the profile of RF amplitude γB1 at the beginning and end of the sweep. For example, using half-sine funeiions, It can be attenuated by shaping. However, from simulations, this species Smoothing requires modifying the offset dependence of the multi-quantum excitation using the series shown in Figure 2. You can conclude. This is probably because the efficiency of the sequence is only relative to the relative RF phase. This is probably because it is sensitive to the relative amplitudes of the three pulses.

The magnitude of the multi-quantum coherence generated by the excitation sandwich in Figure 2 is thus: is largely uniform for all offsets, but the phase is strongly frequency dependent. It is true.

According to the present invention, the phase of the multi-quantum coherence is changed by another chirb pulse. can be re-imaged.

FIG. 8 shows a fourth pulse having twice the sweep speed and half the duration of the preceding pulse. The chirb pulse is re-imaged by the excitation sandwich after a reimaging delay time t. shows the sequence inserted immediately after the code is formed. Two quantum cohere at the point of echo The phase of the spin is thick (independent) from the offset of the two spins involved.

The phase φ4 of this pulse, which determines the phase of the two-quantum echo, can be freely selected. 8th The relative RF amplitude γB0 of the four pulses in the figure is optimized to a ratio of 1:2.8:1:3.3. It was done.

This sequence actually allows a multi-quantum coherence with uniform phase for all offsets. This makes it possible to obtain The dashed line at the bottom of Figure 9 indicates that the two-quantum phase Φ= Itctin((DQC, IF/(DQC),) has little offset dependence. It is confirmed that it is mostly flat.

In Figure 9, offset ΩA/2π = 10kHz ji 20kHz remains constant in the middle of the z-sweep range, but Ω,/2π decreases by 10k towards the right edge. Hz to 20 kHz. The solid line in Figure 9 is the simple 3-pulse line in Figure 2. immediately after the sand inch, i.e. without reimaging (magnitude) and The dashed line represents the echo phase in the sequence with two quantum reimaging pulses in Figure 8. shows the magnitude and phase of time.

Once generated, reimaged, and allowed to unfold during period t1, multi-quantum coherence simply reduces to one quantum coherence observable by a single π/2 chirb pulse cannot be converted.

From such pulses, anti-phase-quantum terms, residual multi-quantum coherence and longitudinal As a result, an offset-dependent mixture of 2 spin order 2 tx'z is obtained. There will be. In view of this difficulty, at the beginning of the detection period t2 of the two-dimensional experiment, the same phase-quantity Two methods seem practical for obtaining child coherence.

The first one is that the multi-quantum coherence increases as a function of offset before the conversion to anti-phase magnetization. involves creating a specific phase distribution of the

The second method relies on the concept of symmetric excitation and detection, and the preparation and detection of multi-quantum experiments. The part has an apparent time-reversal. be aligned to achieve this.

The sequence in Figure 8 shows that two quantum coherences occur at different times during the third pulse. conversely, it becomes possible to reimage the quantum coherence. The phase of the two-quantum coherence prior to the exchange is intentionally expanded (sptead out). ) is also good. In this way, the R and F fields sweep through the spin A and M resonances. The instantaneous RF phase and two-quantum coherence at times tA and ty (see Figure 1) when It can be ensured that there is a suitable relationship between the phase of Shown in Figure 10. This means that the two quanta assumed to consist of pure fDQc), as The coherence is post-echo phase dispersed and the effective flip angle π of duration t,/2, That is, a Chirb reimaging pulse with twice the sweep rate of the pulse of duration t, This can be achieved by applying

A subsequent Chirb π/2 pulse of duration t causes (antiphase) - quantum coherence A conversion to Lens is brought about. The initial RF phase of this pulse is must be selected.

The amplitude (but not the phase) of the observable magnetization thus recovered is the resonant frequency Uniform for all pairs of numbers Ω8 and Ω2.

In layman's terms, the whole process is between the third pulse and the echo in Figure 8. This is the opposite of what happens.

In Figure 10, when the second pulse passes through the resonant frequency, it has two quantum phases. The relationship between the RF phases is 2IAxy and 21AzI”xv (not shown). A transformation is due to take place. These antiphase-quantum terms are then The in-phase terms I□7 and I”XY (also not shown) can be changed, and these At the end of the third period, the anti-phase quantum co An echo of Heerence is formed. All pulses were swept over 20kHz. drawn. The π and π/2 pulses had durations of 8 and 16 ms, and the RF vibration The widths are 1020Hz and 309Hx, respectively, and the initial RF phase of the second pulse is shifted by 246° with respect to the first pulse. Offset is Ω-=8 kHz and ΩM = 12 kHz. The delay time is τ1=8ms, τ2= 54.5 milliseconds, τ3 = 62.5 milliseconds, and in this case, τ3 = τ2 + τ. . The latter two are τ2+τ, + and 3=1. /J AM-JAv=8H at this time It was adjusted to be z.

Figure 11 shows the reconversion system of Figure 10 for pairs of spins with different offsets. Indicates column efficiency. In every spin pair, we have an eco-temporal pure (DQ C) (see pulse sequence in Figure 10). The offset, in contrast, is ΩA For /2π, Ω from Ok Hz to 10kHz above, 20k for /27C The frequency was varied from H2 down to 10 kHz, but all chirb pulses It was swept from 0 to 20 kHz. Pulse duration, RF amplitude, RF phase and The distance and spacing are the same as in FIG.

Thus, for every pair of offsets Ω and Ω−, two lines aligned along the Starting from quantum coherence, the inverse of uniform magnitude over a wide offset range It becomes indeed possible to obtain phase-quantum terms. However, in the series of Figure 10, If a signal is recorded immediately after the second pulse (opposite phase), a strong offset-dependent phase can be detected. It will be distributed. In the ω1 dimension by such an experiment, however, we can obtain It is possible to obtain a pure absorption linearity rather than in the ω2 dimension of the frequency domain. Become.

The second method of detection is 1. symmetrical to modulating populations It is based on the idea that excitation and reconversion can be transferred to the next observable magnetization. It's on.

For example, a chain of transformations in a desired state consisting of multiple quantum coherence If we prepare the system, under certain conditions we can reverse all the previous transformations and find the beginning of the equilibrium magnetization. It is possible to recover the initial state.

According to this concept, the sequence preparation and reconversion part of FIG. and V. Inserting a deployment period t1 between the preparation and reconversion parts By doing so, eos (Ω.

A longitudinal Zeeman magnetization modulated as +Ω&1)tl is obtained.

This is then combined directly with a hard π/2 pulse or, preferably, a Chirb π/2 pulse. observable by a subsequent echo sequence consisting of a Chirb π pulse with twice the sweep rate It may be converted to transverse magnetization.

Propagators U and V each sample three chirped pulses of total duration. Represents doinch. If the expansion period is t1 = 0, and if the non-selective hard If a π pulse is inserted between U and V (vertical arrow), the effect of ■ is generated by U. The resulting two-quantum coherence is divided into two over a wide range of offsets Ω6 and Ω2. It is to convert back to the Zeeman terms rA□ and IM of spins A and M.

The effect of the preparation propagator U is if the condition V = U-1 can be satisfied. , is ``undone'' (or ``reversed in time'') by the action of propagator V. Good too. This means that for the corresponding (Hermitian) Hamiltonian operator Hv = - Corresponds to the requirement conditions H and J. These relations are expressed as follows: Immediately after the reconversion in the two-dimensional experiment shown in Figure 12, the Z component of magnetization 1, =IA, +I town. If we consider the expected value, it can be easily understood. Vezp(-iHoll)　U　I　zU-”exp(iHoll)　V-’I In the L formula, U = ezp(-i 5 Hu(t) d+, and HO is the free age of formula (1) It is a differential motion Hamiltonian. If t, = 0, S = trf(V-” I, V) (U I, U-') (6) is obtained, therefore if U-'=V (ie if Hv=-HU), the maximum possible signal tr(lx2) is obtained.

Since the sign of a complete Hamiltonian cannot be simply inverted in practice, the RF amplitude and position We must explore whether synergistic effects can be achieved.

First, the offset ΩA(+,)=ω, with respect to the time-dependent transmitter frequency, . -ω, 0) corresponding to a chirped pulse acting on an isolated spin Let us consider the following: Based on Figure 12, at contrasting times τ−t′ and +t′, the same One offset and the same RF amplitude must be obtained, in which case the preparation period Corresponding to the end and the beginning of the reconversion period: ΩA(τ−1′) = ΩA(τ++ ’) or ω+tp(τ−1′)=ωRF(τ+1′) and γB! (τ−:′ ) = γB + (τ + t') (8) This means that the direction of the sweep must be changed. It means no. Hamiltonian H7 now associated with propagator V If we choose the phase such that is the complex conjugate of Hl, i.e. if U''=V-1 So that Hv (τ1+’) = H,” (τ-1’) = ΩA (τi) ■: +γB, ( 1; cosφ−1’, sinφ) (9), then + in case of complete time reversal Instead of t(IU-'IzUl2), the signal tr((U-''IzU) It can be shown that 21 is obtained. Regarding the matrix elements, we have: S; Σ11ρ, 2 and 11zU) 1. The two equations differ by the phase factor and for the first one, the maximum U signal strength is achieved over t□. He can only speak deaf words. If there is a so-called spin-reversal transition during the expansion interval, If an associated precession is obtained, an in-phase signal can indeed be obtained. the The transition 11〉〈jl is transformed into its complex conjugate 1j〉〈il by π, the pulse possible, and if we apply such a pulse during deployment, we can obtain the exact time-reversal equivalent. etc. can actually be obtained. In a two-spin system, the two-quantum coherence 1αα〉〈β β1 and zero quantum coherence Iαβ〉〈αβ1 both correspond to spin reversal transition do.

If we introduce the scalar coupling term H1=2πJIA-IM as in equation (1), If so, we can reverse the sign of this term, except perhaps by using a selection pulse, which we want to avoid. It should be noted that there is no obvious way to One solution to this problem One possible solution is to adapt the pulse sequence to the time dependence resulting from the coupling term. Ru. This means that at the end of the preparation period, the coupling has completely transformed the in-phase magnetization into an out-of-phase term. This can be achieved if this coincides with the time when the exchange was completed. Figure 2 3-pulse sandwich inch Since we are relying on the series to create the maximum antiphase term in any case, this condition is Totally naturally satisfied. In the t1 interval, two-quantum and zero-quantum coherence Since these correspond to spin-flip transitions, a hard π pulse is used. The image may be re-imaged.

Of course, this π pulse is different in a series consisting only of chirb pulses. Although it may seem like an unorthodox element, in principle this pulse is The multi-quantum coherence generated under the action of data U is π for all offsets. /1 and has phase φ=0, that is, pH=　(U-1IzU)　, + If is purely real, it may be omitted. For this purpose, the principle As such, the reimaging sequence discussed above can be used. The reason is that it is Two-quantum and zero-quantum coherences with phase common to all offset pairs This is because it causes a However, the multi-quantum coherence along the X-axis We need specific values for the phase and amplitude of all pulses to obtain.

FIG. 13 shows the difference between the proposed time-reversed sequence shown for the case t□=0. It shows the time evolution of the antiphase-quantum term and the magnitude of two-quantum coherence. Re-inversion The phases during the transition period have sign relative to those in the preparatory period to satisfy the condition of equation (9). has been changed.

A stiff π pulse (not shown) is applied between excitation and reconversion.

The RF frequencies of the chirped pulses are π/2 and π/2, respectively, as in FIG. For π pulses RF amplitude γB1 = 309 Hz and 860 Hz, O in 16 zU seconds It is swept up and down between kHz and 20kHz. The resonance frequency is ΩA/2π = 9 kHz and ΩM/2π = 11 kHz, J-M = 8 8! It is. Almost all of the initial Zeeman order was converted to two-electron coherence. Be careful not to get caught.

Figure 14 shows that the longitudinal magnetization profile obtained at the end of the series is mostly offset. Indicates independence. The performance increases with increasing sweep speed, as in Figure 7. It tends to improve with time.

The Zeeman terms ＾ and IM □ reconverted from the two-quantum coherence are ΩA/2 For π, from Ok Hz up to 20 kHz, for Ω&l/2π, 40 The offset function is varied symmetrically from kHz down to 20kHz. Shown as a number. The solid lines are R for π/2 pulse and π pulse, respectively. F amplitude γBt=618Hz and 1720H! and the phase φ3=122', 8 mi The response to a chirped pulse swept between 0 and 40 kHz in seconds is shown.

The dashed line is 7B1=440Hz and 1220H! and 16 at φ3=526 Corresponds to a 40kHz sweep of zU seconds.

The final path to observable one-quantum coherence of Iz generated by the propagator V A conversion is a chirb π/2 pulse followed by a half of the chirb π/2 pulse. and an amplitude of 2 to 4 times, preferably 2.8 times. This may be achieved by an echo sequence consisting of .

Bruker A equipped with an Oxford Research Systems selective excitation device Experiments were performed on an M-400 spectrometer. Chirped pulses have a phase that is a quadratic function of time, That is, φ=1/2b (t-t,) 2 (where to corresponds to the center of the pulse and b is the sweep speed) Files created by a Pascal program running on a computer are Br1I It is converted by the ker 5HAPE package and stored in the waveform memory (wave (otm). me+aoty) was created.

The two-quantum experiment was performed on the proton system. These experiments were performed using the reimaging method shown in Figure 8. Combining octopus quantum excitation with an eco-simultaneously “hard” π/2 monitoring pulse It is. Figure 15 shows two cases with JAM = 5Hz and JA'M = 15Hx. contains two subsystems (subs7s+s+u) with scalar bonded protons of Chloride (IR5゜2SR,4R3,5R3,6R3)-exo-(2-nitrophe nylthio)spiro[bicyclo[2,2,21octane-2゜2'-oxiranoco -5-endo-yl spectrum is shown.

Since these couplings are multiples of each other, the delays in the experiment are for both subsystems. can be optimized at the same time. The chirp pulse has a duration t,; 8 ms. , has a τ delay of 42 ms, so the total duration of the excitation sandwich is 100 milliseconds χ(2JAM)−”χ3(2JA·2·)−1.

Due to storage capacity limitations, only 256 time intervals can be defined for each pulse. Therefore, the amount of phase increase is determined by frequency sweep with a window of ±5 kHz on both sides of the carrier frequency ( window). First and last time within each pulse For the interval, the phase increase was 0.98 rad. This is the simulation was used to optimize RF amplitude and phase under relatively rough numerical conditions. . The optimal amplitudes γB1 of the four pulses in FIG. 8 are 309Hx and 860Hx, respectively! h, 309B! and 1020H! It was found that the last two pals The phase of the phase was shifted by 150” with respect to the first two. Good results calibrate the amplitude γ of a single chirb pulse for maximum generation of transverse magnetization and the amplitude of the other pulses in proportion to the optimized value, i.e. 2.8:1:3.3. Obtained by setting. Cycling the phase of the receiver with a hard reconversion pulse This suppresses undesirable signals due to zero quantum and -quantum coherence. I was able to control it.

The spectrum in Figure 15 has frequencies ω1 = (Ω6 + Ωv) and ω2 = Ω6 and Ω a The skew (skev) is arranged symmetrically with respect to the two quantum diagonal ω, = 2ω2. Shows a pair of multiplets. The cross section is double for each of the two subsystems. The opposite phase upper/lower characteristics of the front are clearly revealed.

Selected cross sections taken parallel to the ω2 axis show the antiphase structure of the doublet. vinegar The spectrum consists of a four-pulse channel sequence using two-quantum reimaging in Figure 8, and a subsequent using a non-selective “hard” π/2 pulse for reconversion to antiphase-quantum coherence. I got it. The spectral width is 10Hz and ±5Hz at ω1 and ω2. Yes, the data matrix is 512 x 4 before zero filling and IK after 1 It was XIK. The chirb pulse was generated by phase modulation of the RF carrier. .

The performance of the method of the invention, i.e. the quantum coherence due to the first sequence of chirped RF pulses. multi-quantum coherence by excitation of the lenses and a second series of Chirb RF pulses. Devices that allow reconversion to magnetization are sensitive to large offsets. Original transformation spectroscopy (“NOESY”), correlation spectroscopy (rcO3YJ), heteronuclear and NMR-imaging experiments can also be performed in spectrometers adapted to perform homonuclear correlation spectroscopy. , especially in NMR-in-vivo spectroscopy. Such devices also Book 5! belongs to the subject matter of the present invention and is intended to be included in the scope of the claims. .

The frequency-modulated chirb pulse sequence proposed by the present invention is a high-field carbon-13 A two-quantum component is applied to a system with a wide spectral range, as actually occurs in the case of spin pairs. It holds considerable promise for expanding optical methods. The method of the invention applies to all kinds of nuclear magnetic systems. It is understood that it is applicable to

Substantial advances can be made by using frequency modulation systematically. To the present invention The Chirb pulse sequence according to Applicable to dimensional experiments, systems with quadrupolar coupling in anisotropic phase, and dipolar coupled systems. Ru.

Table 1 12 I:yI? ’=　[(2x:t:)”10(2x’, x’:)2]”12 I gradient’ I= [(2I”I”) 2+(2IIIτ) 21 L/21 Xγ l ZQCl = [(ZQ(j, +(ZQ(j, koIDQC+=[( DQC), + fDQ (j Table 1 Size of knee quantum and two quantum coherence Definitions of shorthand notations used in , symbols refer to the (real) expected value of the corresponding operator. It is something to do.

Fig, 1 Fig, 3b Fig, 4a Fig, 9b Fig, 14 Copy and translation of amendment) Submission (Article 184-8 of the Patent Law) January 11, 1993 U −

Claims (24)

[Claims] 1. A first π/2 pulse is applied to the nuclear spin system to which a magnetic field is applied. The first π pattern is generated after the elapse of the defocusing time interval τ following the first π pattern. pulse and another time interval τ following the first π pulse. The first of the RF pulses comprises an elementary sandwich consisting of a second π/2 pulse In NMR pulse experiments, multiple quantum, especially two-quantum excite coherence, and transfer the excited multi-quantum coherence to the nuclear spin system with R. By irradiating with a second series of F pulses, the magnetization is reconverted to longitudinal magnetization and free induction decays. In the method of inducing The three RF pulses of the elementary sandwich each have a duration tp. pulse, during which the frequency of the excitation RF field is equal to the lower limit frequency ωRP. From min to the upper limit frequency ωRFmax or vice versa, with a monotonous relationship with respect to time. a swept chirped pulse with a duration tp/2 and a fourth chirped pulse having an amplitude 2 to 4 times closer to the amplitude of the first π/2 pulse; A method characterized by following. 2. said sandwich is immediately followed by said fourth chirped pulse. The method according to claim 1. 3. The amplitude of the first π pulse in the basic sandwich is equal to the first π/ 2 to 4 times, preferably 2.8 times, the amplitude of the basic sandwich. The amplitude of the second π/2 pulse is approximately the same as the amplitude of the first π/2 pulse. A method according to claim 1 or 2, characterized in that: 4. The amplitude of the fourth chirped pulse is equal to the amplitude of the first chirped pulse in the basic sandwich. The amplitude of the preceding claims is approximately 3.3 times the amplitude of the π/2 pulse of The method described in any one of the following paragraphs. 5. A first π/2 pulse is applied to a nuclear spin system subjected to a magnetic field with a high magnetic field strength. The first π/2 pulse is generated after the passage of another time interval τ following one π/2 pulse. occurs after one π-pulse and another time interval τ following the first π-pulse. an RF pattern comprising a basic sandwich consisting of a second π/2 pulse the first series of russes, in particular the first series of In NMR pulse experiments, multiple quantum, especially two-quantum coherence can be detected by irradiating a series of The excited multi-quantum coherence is transmitted to the nuclear spin system by RF. By irradiating with a second series of pulses, the magnetization is reconverted to longitudinal and free induction decays. In a method of inducing spin echoes generated by a first train of RF pulses, A second system of RF pulses after a subsequent phase dephasing time interval τ1 The train is started and the second train of RF pulses is such that the frequency of the RF field reaches the lower limit frequency ωRP From MIN to upper limit frequency ωRFMAX or vice versa, with a monotonous relationship with respect to time. Chirb reimaging π-pulse with swept duration tp/2 and duration tp a subsequent chirped π/2 pulse. 6. The second series of RF pulses further includes a delay following said subsequent chirped π/2 pulse. Another chirp π pattern of duration τx=tp/2 applied after the extension time τ2 the other chirped π pulse. Claim 5 characterized in that the detection is performed after a final delay time τ3 following the Method described. 7. The delay time τ2 is set such that τ2≈0, and the final delay time τ3 is selected such that τ3≒τπ. the method of. 8. The delay time τ2 is chosen such that τ2≒(2JAM)-1, and the previous The final delay time τ3 is chosen such that τ2+τπ+τ3≒JAM−1, and JA characterized in that M represents the scalar coupling constant of the coupling between spin A and spin M; The method according to claim 6. 9. The delay time τ2 is chosen such that τ2≈(4JAM)-1, and The final delay time τ3 is selected so that τ2+τπ+τ3≈(2JAM)−1. is selected, and JAM represents the scalar coupling constant of the coupling between spin A and spin M. 7. A method according to claim 6, characterized in that: 10. The amplitudes of the two chirped π pulses of said second series of RF pulses are The amplitude is 2 to 4 times, preferably about 2.8 times, the amplitude of the π/2 pulse. The method according to any one of claims 6 to 9. 11. A first π/2 pulse is applied to a nuclear spin system subjected to a magnetic field with a high magnetic field strength. Occurs after the elapse of the defocusing time interval τ following one π/2 pulse. after the elapse of a first π-pulse and another time interval τ following the first π-pulse. comprising an elementary sandwich consisting of a second π/2 pulse generated at Multi-quantum , in particular, to excite the two-quantum coherence and to convert the excited multi-quantum coherence into the By irradiating the nuclear spin system with a second series of RF pulses, it is reconverted to longitudinal magnetization. , in a method of inducing free induction damping, The three RF pulses of the elementary sandwich each have a duration tp. pulse, during which the frequency of the excitation RF field is equal to the lower limit frequency ωRF From min to the upper limit frequency ωRFmax or vice versa, with a monotonous relationship with respect to time. The chirped pulse is swept, and after the development time τ1, the basic sandwich In addition, precisely time-reversed sandwiches, each having a duration tp and each other The upper limit frequency ωRFmax to the lower limit frequency ωR are separated by a time interval τ, respectively. Three chips of type π/2, π and again π/2 swept to Fmin or vice versa. A method characterized by a series of sharp pulses. 12. The amplitude of the first π pulse in the basic sandwich is equal to the amplitude of the first π pulse. 2 to 4 times, preferably 2.8 times, the amplitude of the basic sandwich. The amplitude of the second π/2 pulse of the first π/2 pulse is approximately the same as the amplitude of the first π/2 pulse. 12. The method according to claim 11, characterized in that: 13. 13. The method according to claim 11 or 12, characterized in that t1≈0. 14. At either the beginning or the end of the deployment time interval t1, a hard ) A reimaging π pulse is applied. The method described in section. 15. During the development time interval t1, after the basic sandwich a duration tp/ 2 and a fourth chirp pattern having an amplitude of 2 to 4 times the amplitude of the first π/2 pulse. 13. A method according to claim 11 or claim 12, characterized in that the pulse continues. 16. The basic sandwich is immediately followed by the fourth chirped pulse; The amplitude of the chirped pulse of 4 is equal to the first π/2 pulse in the elementary sandwich. 16. The method of claim 15, wherein the amplitude is approximately 3.3 times the amplitude of . 17. The second sequence of RF pulses includes a chirped π/2 pulse followed by a half the duration of the preceding chirp π/2 pulse and 2 to 4 times, preferably about 2. Claim 1, characterized in that it consists of a chirped π pulse with an amplitude of 8 times The method according to any one of items 1 to 16. 18. Acquisition of a free induction decay (FID) signal follows the second train of RF pulses. The acquired signal data is evaluated using Fourier Transform (FT) techniques. A method according to any one of the preceding claims. 19. The frequency of the excitation RF pulse has a linear relationship with time, with the lower limit frequency increasing. The preceding claims are characterized in that they are swept up to the limiting frequency or vice versa. The method according to any one of the above. 20. Two-dimensional light conversion spectroscopy any one of the preceding claims applying to The method described in. 21. Any one of claims 1 to 19 applied to homonuclear correlation spectroscopy Method described. 22. Any one of claims 1 to 19 applied to heteronuclear correlation spectroscopy Method described. 23. NMR imaging and/or NMR-in-vivo 20. A method according to any one of claims 1 to 19 applied to spectroscopy. 24. NMR spectrometer, in particular the method according to any one of the preceding claims for exciting transverse magnetization in NMR experiments. Frequencies designed to allow fast frequency sweeps of the RF electromagnetic field used a modulator, wherein the frequency swept excitation field has a sweep rate, an RF amplitude, a sweep reaction; Frequency Modulator Adjustable with respect to Sweep Repetition and further, an excitation field frequency range is defined by frequency modulation. designed to allow adjustment of the phase shift of the excitation field with respect to the carrier frequency. phase controller and a phase-sensitive receiver input signal and RF carrier frequency. The phase shift between the receiver reference signal and the receiver reference signal is maintained in a coherent phase relationship with respect to A phase shift device that allows adjustment of the pitch and an RF pattern for multi-quantum coherence excitation. transverse the excited multi-quantum coherence with the first series of pulses, especially the chirped pulses. Generation of RF pulses to reconvert magnetization, especially with a second series of chirped pulses and a time control device that controls the An NMR spectrometer comprising: