Magnetic Resonance in Food Science: Defining Food by Magnetic Resonance: Volume 349 (Special Publications) - Hardcover

Hardcover

ISBN 10: 1782620311 ISBN 13: 9781782620310

Publisher: Royal Society of Chemistry, 2015

Synopsis

Capturing contributions from a whole range of applications in food and representing the latest technical innovations, this will be a contemporary book on the topic aimed at academics and industrialists who are committed to the utilisation of MR tools to improve our understanding of food.

About this title

From the Back Cover

Magnetic Resonance has become an established technique to improve the understanding of food systems. Capturing contributions from a whole range of applications in food and representing the latest technical innovations, this will be a contemporary book on the topic. Based on a conference which has established an international reputation as the forum for advances in applications of magnetic resonance to food, the coverage will be dedicated to multiscale definition of food, quantitative NMR (qNMR), foodomics, on-line non-invasive NMR (dedicated to Brian P. Hills), quality and safety and new developments in the area. It is aimed at academics and industrialists who are committed to the utilisation of MR tools to improve our understanding of food.

Excerpt. © Reprinted by permission. All rights reserved.

Magnetic Resonance in Food Science

Defining Food by Magnetic Resonance

By Francesco Capozzi, Luca Laghi, Peter S. Belton

The Royal Society of Chemistry

Copyright © 2015 The Royal Society of Chemistry
All rights reserved.

ISBN: 978-1-78262-031-0

Contents

Rapid determination of food quality using steady state free precession sequences in TD-MNR spectroscopy L.A. Colnago, T.B. Moraes, T. Monaretto, F.D. Andrade, 1, 
Quantitative NMR, 
60 MHz 1H NMR spectroscopy of triglyceride mixtures A. Gerdova, M. Defernez, W. Jakes, E. Limer, C. McCallum, K. Nott, T. Parker, N. Rigby, A. Sagidullin, A. D. Watson, D. Williamson, and E. K. Kemsley, 19, 
Usefulness of 1H NMR to study the food lipolysis during in vitro digestion B. Nieva-Echevarría, E. Goicoechea, M.J. Manzanos and M.D. Guillén, 31, 
Quantitative NMR assessment of polysaccharides in complex food matrices E.J.J. van Velzen, S. Dauwan, N. de Rool, C.H. Grün, Y. Westphal, and J.P.M. van Duynhoven, 40, 
Quality and Safety, 
Magnetic Resonance analysis of dairy processing suitable tools for the dairy industry R. Anedda, 51, 
NMR spectroscopic studies in saffron authenticity and quality (within the frame saffronomics cost action FA1101) R. Consonni, L. R. Cagliani, M. G. Polissiou, E. A. Petrakis, M. Z. Tsimidou, S. Ordoudi, 65, 
Food NMR optimized for industrial use-an NMR platform concept E. Humpfer, B. Schütz, F. Fang, C. Cannet, M. Mörtter, H. Schäfer, and M. Spraul, 77, 
A new ultra rapid screening method for olive oil health claim evaluation using selective pulse NMR spectroscopy E. Mellioul, P. Magiatis and K.B. Killday, 84, 
Profile of the positional distribution of fatty acids in the triacylglycerols as an index of quality for palm oil (or any oil or fat) S. Ng, 93, 
On-line Non-invasive NMR, 
1H-NMR relaxometry and imaging to assess fat content on intact pork loins V. Bortolotti, P. Fantazzini, C. Schivazappa, M. Vannini, E. M. Vasini, R. Virgili, 101, 
Multiscale Definition of Food, 
19F labelled polyion micelles as diffusional nanoprobes D.W. de Kort, F.J.M. Hoeben, H.M. Janssen, N. Bourouina, J. Mieke Kleijn, J.P.M. van Duynhoven and H.V. As, 111, 
Double emulsion character with PFG-NMR- methods: WOW and OWO R. Bernewitz, E. Caro, D. Topgaard, H.P. Schuchmann, G. Guthausen, 120, 
Assessment of TD-NMR and quantitative MRI methods to investigate the apple transformation processes used in the cider-making technology C. Rondeau-Mouro, S. Deslis, S. Quellec, R. Bauduin, 127, 
Foodomics, 
A 1H NMR-based metabolomics approach on dietary biomarker research in human urine A. Trimigno, G. Picone, F. Capozzi, 143, 
1H NMR metabolic profiling of apulian EVOOs: fine pedoclimatic influences in Salento cultivars L. Del Coco, S.A. De Pascali, F.P. Fanizzi, 154, 
Addition of essential oils to cows' feed alters the milk metabolome-NMR spectroscopic studies of "nature's perfect food" U.K. Sundekilde, M.R. Clausen, J. Lejonklev, M.R. Weisbjerg, M.K. Larsen, and H.C. Bertram, 161, 
High-resolution magic angle spinning studies of semi-hard Danbo (30+) cheese-impact of processing condition and relation to sensory perception S. Lamichhane, C.C. Yde, L.H. Mielby, U. Kidmose, J.R. Møller, M. Hammershøj and H.C. Bertram, 171, 
Changes in the 1H NMR metabolic profiling of mussels (Mytilus galloprovincialis) with storage at 0°C V. Aru, M.B. Pisano, P. Scano, S. Cosentino and F.C. Marincola, 181, 
Applications of 1H-NMR metabolomics: from individual fingerprints to food analysis A. Luchinat and L. Tenori, 190, 
New Developments, 
Compost biodegradation by 1H magnetic resonance and quantitative relaxation tomography V. Bortolotti, P. Fantazzini, M. Vannini and E.M. Vasini, 203, 
1H NMR spectroscopy of lipoproteins-when size matters F. Savorani and S.B. Engelsen, 211, 
Subject Index, 224, 

CHAPTER 1

RAPID DETERMINATION OF FOOD QUALITY USING STEADY STATE FREE PRECESSION SEQUENCES IN TD-NMR SPECTROSCOPY

L.A. Colnago, T.B. Moraes, T. Monaretto, F.D. Andrade

1 INTRODUCTION

The use of time-domain NMR spectroscopy (TD-NMR) in food science began more than 40 years ago with the introduction of the small benchtop NMR analyzer. Since then, TD-NMR has become one of the most robust, rapid, cost-effective and versatile tools in the food industry. Earlier TD-NMR applications were primarily based on quantitative analysis using the intensity of free induction decay (FID) and/or spin echo signals. In the last two decades, the use of relaxometry and/or diffusometry methods have expanded the application TD-NMR in food science exponentially.

The majority of these applications use the Carr-Purcell-Meiboom-Gill (CPMG) pulse sequence. This sequence is very robust, rapid and yields an exponential decay that is dependent upon the transverse relaxation time (T2). Therefore, CPMG has been used as an all-purpose sequence in TD-NMR applications and is a standard pulse sequence present in commercial and homemade TD-NMR spectrometers. CPMG has been used to study food products such as oilseeds, fresh meat, fish, and fruit, as well as industrialized and packaged food products.

The longitudinal relaxation time (T1) measurements using inversion-recovery (IR) or progressive saturation pulse sequences have rarely been used in food analysis due to the length of experiment time. Pulsed field gradient spin-echo (PFGSE) pulse sequences are the second most used pulse sequence in TD-NMR applications. PFGSE has been used to measure the water self-diffusion coefficient, water mobility, and droplet size in several food products. However, PFGSE requires an additional hardware accessory that is not available for all TD-NMR spectrometers. Thus, there is an effort towards the development and implementation of rapid TD-NMR analytical methods that meet the growing demand for tools of quality assessment.

Accordingly, we have been developing steady-state free precession (SSFP) pulse sequences for TD-NMR spectroscopy since 2000. SSFP sequences have been used in quantitative analysis similarly to analyses performed with FID or spin echo. However, the signal-to-noise ratio (SNR) with SSFP is much higher than that obtained with FID or echo in the same average time. Moreover, SSFP sequences can also be used in fast flow (online) quantitative measurements of liquid or solid samples. The theory for quantitative analysis using the amplitude of an SSFP signal is presented in section 2.1.

Further advantages of SSFP sequences are: the dependence of the transient signals on two relaxation times (T1 and T2), the data are collected in a length of time similar to CPMG and it does not require special hardware and therefore can be implemented on any modern TD-NMR spectrometer. The theory for the evolution of the NMR signal submitted to a train of pulses (SSFP sequence) is presented in section 2.2.

2 THEORY

2.1 Amplitude of the NMR signal in the SSFP regime

SSFP sequences have been used to improve the SNR in pulsed NMR spectroscopy since 1958. It is a simple pulse sequence consisting of a train of radiofrequency pulses (rƒ) with the same phase and flip angle (θ), and the time between pulses (Tp) is shorter than T2 (Tp< T2) (Figure 1).

In 1966, Ernst and Anderson derived the analytical solution for the SSFP regime. They showed that the SSFP signal is composed of FID and echo signals. The echo component (M-) immediately preceding the pulse is given by equations 1 through 3, and the FID (M+) component is given by equations 4 through 6.

M-x = M0 (1 - E1) [E2 sinθ sinΦ]/D (1)

M-y = M0 (1 - E1) [E2 sinθ cosΦ] - E22sinθ/D (2)

M-z = M0 (1 - E1) [1- E2 cosΦ sinΦ] - E2 cosθ(cosθ - E2)]/D (3)

M+x = M-x (4)

M+y = M0(1 - E1) [1 - E2 cosΦ)sinθ]/D (5)

M+z = M0(1 - E1)[E2 (E2cosΦ) + (1 - E2 cos Φ) cosθ]/D (6)

where D = [(1 - E1cosθ)(1 - E2cosΦ)] - [(E1 - cosθ)(E2 - cosΦ)E2], with the precession angle Φ = Ωt, offset frequency Ω = wref - w0, and relaxation components E1 = exp(-Tp/T1) and E2 = exp(-Tp/T2).

With these equations it is possible to calculate the magnitude of the magnetization in the xy plane after the nth rƒ pulse, assuming Tp<< T1.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (7)

Therefore, the amplitude of the SSFP signal is dependent upon the flip angle θ, precession angle Φ = ΩTp and T1/T2 ratio. The magnetization goes to null |M| [arrow right] 0, when [square root of (2 - 2 cosΦ [arrow right] 0)] or

Φ = n2π (8)

where n is an integer.

Figure 2 shows the dependence of the NMR signal amplitude upon the precession angle Φ and frequency offset for θ = 45° and 90°, Tp = 0.3 ms, T1 = 150 ms and T2 = 50 ms, according to equation 7.

For Φ = n2π, the magnetization is minimal because the FID and echo components are dephased by 180°, resulting in destructive interference. For Φ = (2n + 1)π, the FID and echo are in phase and the constructive interaction creates a maximum signal intensity when θ = 90°.

Equations 1 through 6 show that the behavior of the magnetization in the SSFP regime is complex and depends on a series of experimental parameters, such as Φ, θ, Ω, Tp, T1 and T2. However, these analytical descriptions do not include the effect of other parameters on the SSFP signal, such as Tp variation and phase alternation.

To fully describe the SSFP phenomenon we have numerically simulated (Matlab) the influence of the all the above parameters based on the rotation matrix and the method of the sum of isocromats, in which the Lorentzian distribution was assumed.

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (9)

and

1/T*2 = 1/T2 + 1/T2inom

where T2inom = 1/γΔB0.

Figure 3 shows the numerical simulations for the TD-NMR signals after reaching the steady state regime. The time necessary to reach the steady state is discussed in section 2.2. The steady state signals were simulated using T1 = 100 ms, T2 = 50 ms, T2* = 0.5 ms, θ = 90°, a frequency offset of 8.333 (Figures 3A to D) and 6.666 (Figure 3E) KHz and various Tp values. In Figure 3, the pulse is observed in the center of the window (t = 0). The FID component after the pulse is on the right side of t = 0, and the echo component is on the left side of the pulse.

Figure 3A shows the NMR signal for Tp = 5T1. This figure shows an FID signal with maximum amplitude. With this pulse repetition rate, the echo signals are not observed. Figure 3B shows the NMR signal for Tp = T2. In this condition the NMR signal is in the SSFP regime and is composed of an FID and an echo between the pulses. The FID signal has higher amplitude than the echo signal. The FID amplitude in Figure 3B is lower than that of the FID in Figure 3A because Tp< 5T1, which does not allow the return of the magnetization to thermal equilibrium.

Figures 3C to E depicts more than one period between the pulses, in the interval of -1.5 to 1.5 ms. Figure 3C shows two periods for Tp = 2.9T2* in which the FID and echo signals have similar amplitudes, and the FID decays faster than T2* compared to the FID decay in Figures 3A and B. This faster decay is due to the partial destructive interaction between the FID and echo in the center of the SSFP signals.

Figures 3D and E show the SSFP signals for Tp = 0.3 ms < T2*. In this condition the overlap between the FID and the echo signal is maximal, yielding a special SSFP regime, known as Continuous Wave Free Precession (CWFP). The amplitude of CWFP signal is strongly dependent upon Φ = ΩTp, as shown in Figure 2. Figure 3D depicts the maximum CWFP signal when Φ = 22π with a frequency offset of 8.333 KHz (constructive interference) and Figure 3E depicts the minimal CWFP signal when Φ = 44π with a frequency offset of 6.666 KHz (destructive interaction).

According to equation 11, the magnitude of the CWFP signal when θ = 90° and Φ = 5π is dependent upon the T1/T2 ratio.

|Mss| = M0/1 + T1/T2 (11)

The magnitude of the CWFP signal is not dependent on the pulse repetition rate, as in conventional pulse sequences (Figure 3A and B). Instead, it depends on the T1/T2 ratio (equation 11), and the repetition time can be short (Tp<< T1, T2< T2*) and without saturation (Figure 3D). Therefore, thousands of CWFP signals can be averaged during one T1 period, thereby enhancing the SNR by one order of magnitude in the same average time used for FID or echo signals. The magnitude of CWFP signals has been used in quantitative analysis, in conventional benchtop spectrometers and in online measurements using long Halbach and superconducting magnets.

2.2 Transient SSFP signal

The evolution of the NMR signal, submitted to a train of pulses (SSFP sequence) has been used in several applications in TD-NMR. In 1977, Kronenbitter and Schwenk proposed the use of the transient SSFP signal to measure T1 and T2. The method consists of two steps: First, the measurements of the T1/T2 ratio, by measuring the maximum amplitude of the SSFP signal as a function of the flip angle (θ) to obtain the optimum θ (θopt), and second, the use of θopt to measure the time constant (T*), equation 12, for the evolution of the SSFP signal, yielding the T1+T2 value. With these two measurements it is possible to determine both relaxation times.

T* = 2T1T2/T1(1 - cosθ) + T2(1 + cosθ) (12)

In 2006, Venâncio T. et al reported that is not necessary to use two steps to measure both relaxation times when the transient SSFP signal is obtained with θ = 90° and Φ = (2n + 1)π.

Figure 4 shows the evolution of the magnitude of the CWFP signal from the first pulse to the stationary regime (|Mss|). This signal undergoes two transient regimes before the steady state is reached. The first transient regime (dark grey) shows an alternation of the amplitude between even and odd pulses followed by signal decay, with the time constant of T2*. When the alternations subside, the signal reaches a quasi-stationary state (light grey). The decay of the quasi-stationary state to the stationary state (white) is governed by the time constant T*. For θ = 90°, equation 12 is reduced to:

T* = 2T1T2/T1 + T2 (13)

Upon rewriting equations 11 and 13, we obtain the following:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII] (14)

Therefore, upon measuring the magnitude of the signal after the first pulse |M0, the magnitude of the CWFP signal |Mss and T*, it is possible to calculate the relaxation times in a single scan experiment using equation 14. The T* value is calculated by fitting the T* decay with an exponential decay function. The T1 and T2 values are obtained with single CWFP experiments are similar to those obtained by Inversion recovery (T1) and CPMG (T2) pulse sequences.

When T1 ~ T2 there is only a small difference in amplitude between the quasi-stationary state and the stationary state of the CWFP signal. This might yields, a T* with large error when the CWFP signal has low SNR.

To solve this problem we proposed the use of a Carr-Purcell sequence, using 90°-refocusing pulses, also known as CP-CWFP (Figure 5). The only difference between CWFP and CP-CWFP sequences is the addition of a pulse, which separates the CWFP pulse train (Figure 1) by the time interval Tp/2. The effect of this modification is shown in Figure 6. The CP-CWFP signal intensity decays to a minimum value (quasi-stationary state) and then increases to the same amplitude observed in the CWFP regime. As shown in Figure 6A, the amplitude variation during T* is much more pronounced in CP-CWFP than in the CWFP sequence. This results in improved fitting of T* for a sample when T1 ~ T2. Conversely, CP-CWFP signals yield a small difference in amplitude during T* when T1 >> T2 (Figure 6B). In this case, T* of CWFP can be fitted with minimal error. When T1 > T2 (Figure 6C) CWFP and CP-CWFP have similar amplitude variations during T*.

(Continues...)

Excerpted from Magnetic Resonance in Food Science by Francesco Capozzi, Luca Laghi, Peter S. Belton. Copyright © 2015 The Royal Society of Chemistry. Excerpted by permission of The Royal Society of Chemistry.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.

Edition details

• Publisher: Royal Society of Chemistry
• Publication date: 2015
• Language: English
• ISBN 10: 1782620311
• ISBN 13: 9781782620310
• Binding: Hardcover
• Number of pages: 232
• Editor: Francesco Capozzi