*3.3.2. Thermal conductivity of water-in-PAO nanoemulsion fluids* 1 Scattering Intensity I (cm-1)

10

100

Figure 13 shows the thermal conductivity enhancement in water-in-PAO nanoemulsion fluids as a function of the loading of water from 0.47 % to 8.6 vol. %, in which the thermal conductivity linearly increases with higher water volume fraction and reaches a maximum of 16 % increase at 8.6 vol. % water. 0.01 0.1 Wave Vector q ( A-1) 12

7.8 vol% 3-Region G-P Fit

10.3 vol% 3-region G-P Fit

4.5 vol% 3-Region G-P Fit

7.8 vol%

10.3 vol%

4.5 vol%

<sup>13</sup> **Figure 13.** Thermal conductivity of water-in-PAO nanoemulsion fluids versus water volume fraction. The linear curve fit is shown for comparison.

### *3.3.3. Viscosity of water-in-PAO nanoemulsion fluids*

The dynamic viscosity of water-in-PAO nanoemulsion fluids with different water volumetric concentrations is shown in Figure 14. All the water-in-PAO nanoemulsion fluids exhibit a shear-independent characteristic of Newtonian fluids. One unique phenomenon that can be seen in Figure 14 is that there is a maximum value in viscosity: it first increases with water concentration, reaches a maximum at 5.3 vol. %, and then decreases. This trend is different from the thermal conductivity shown in Figure 13 and the viscosity trend as observed in other nanoemulsion fluid systems. The maximum in viscosity can be attributed to the attraction force between droplets within the nanoemulsion fluids. The surfactant molecules become hydrated when more water is added inside and their counter ions are released into water which makes surfactants molecules and droplets charged oppositely so that the interdroplet attraction keeps increasing until the hydration process is complete. This may lead to a maximum viscosity in water-in-PAO nanoemulsion fluids as shown in Figure 14. It also coincides with the nonlinear inner structure change with increasing water concentration as seen in Figure 10.

### *3.3.4. Effective heat capacity of water-in-PAO nanoemulsion fluids*

Similarly, the heat capacity of water-in-PAO nanoemulsion fluids was also investigated (shown in Figure 15). As shown in water-in-FC-72 nanoemulsion fluids, the water inside the water-in-PAO nanoemulsion fluids can increase the system's effective heat capacity through

**Figure 14.** Dynamic viscosity of water-in-PAO nanoemulsion fluids versus volume fraction of water.

*3.3.2. Thermal conductivity of water-in-PAO nanoemulsion fluids*

1

**0.14**

*3.3.3. Viscosity of water-in-PAO nanoemulsion fluids*

**0.15**

**0.16**

**Thermal Conductivity** *(W/mK)*

**0.17**

10

Scattering Intensity I (cm-1)

12

100

at 8.6 vol. % water.

96 Neutron Scattering

fit is shown for comparison.

Figure 13 shows the thermal conductivity enhancement in water-in-PAO nanoemulsion fluids as a function of the loading of water from 0.47 % to 8.6 vol. %, in which the thermal conductivity linearly increases with higher water volume fraction and reaches a maximum of 16 % increase

**Experimental Data**

0.01 0.1

Wave Vector q ( A-1)

**0.0 3.0 6.0 9.0**

**Water Volume Fraction (vol%)**

<sup>13</sup> **Figure 13.** Thermal conductivity of water-in-PAO nanoemulsion fluids versus water volume fraction. The linear curve

The dynamic viscosity of water-in-PAO nanoemulsion fluids with different water volumetric concentrations is shown in Figure 14. All the water-in-PAO nanoemulsion fluids exhibit a shear-independent characteristic of Newtonian fluids. One unique phenomenon that can be seen in Figure 14 is that there is a maximum value in viscosity: it first increases with water concentration, reaches a maximum at 5.3 vol. %, and then decreases. This trend is different from the thermal conductivity shown in Figure 13 and the viscosity trend as observed in other nanoemulsion fluid systems. The maximum in viscosity can be attributed to the attraction force between droplets within the nanoemulsion fluids. The surfactant molecules become hydrated when more water is added inside and their counter ions are released into water which makes surfactants molecules and droplets charged oppositely so that the interdroplet attraction keeps increasing until the hydration process is complete. This may lead to a maximum viscosity in water-in-PAO nanoemulsion fluids as shown in Figure 14. It also coincides with the nonlinear

inner structure change with increasing water concentration as seen in Figure 10.

Similarly, the heat capacity of water-in-PAO nanoemulsion fluids was also investigated (shown in Figure 15). As shown in water-in-FC-72 nanoemulsion fluids, the water inside the water-in-PAO nanoemulsion fluids can increase the system's effective heat capacity through

*3.3.4. Effective heat capacity of water-in-PAO nanoemulsion fluids*

7.8 vol% 3-Region G-P Fit

10.3 vol% 3-region G-P Fit

4.5 vol% 3-Region G-P Fit

7.8 vol%

10.3 vol%

4.5 vol%

the high specific heat of water (i.e. *C*water=4.2 J/g C, *C*PAO=1.88 J/g C) and/or the latent heat of water (Δ*H* =334 J / g), depending on the operating temperature of the fluids.

DSC cyclic curves of water-in-PAO nanoemulsion fluids under different water loadings are shown in Figure 15. During the heating and cooling cycles, water nanodroplets undergo a melting–freezing transition in the nanoemulsion fluids. Interestingly, the presence of a single freezing peak in Figure 15 indicates a correspondence of the structural change with increasing water concentration (or water to surfactant molar ratio) as observed in our previous SANS measurement result: there is no obvious melting/freezing peak for water concentrations less than 4.5 vol. %, while the exothermic crystallization peak starts at around −20 <sup>o</sup> C when water concentration is higher or equal to 4.5 vol. %. When the water concentration is increased further above 8.6 vol. %, the freezing peak shifts to effect lower supercooling and peak values.

−20 oC, corresponding to the freezing of water nanodroplets, while endothermal peaks are observed at 0 oC, corresponding to the melting of water nanodroplets. To gain further insight into that, the specific heat of each sample is also calculated and summarized here in Table 2. The calculated *H*f values of water-in-PAO nanoemulsions for different water concentrations from 5.3 to 8.6 vol. % are 17.7 and **Figure 15.** DSC cyclic curves of water-in-PAO nanoemulsion fluids for different water concentrations. Exothermal peaks are observed at −20 <sup>o</sup> C, corresponding to the freezing of water nanodroplets, while endothermal peaks are ob‐ served at 0 <sup>o</sup> C, corresponding to the melting of water nanodroplets.

loading based on the SANS measurement.

Table 2. Specific heat of water-in-PAO nanoemulsion fluids

Figure 15. DSC cyclic curves of water-in-PAO nanoemulsion fluids for different water concentrations. Exothermal peaks are observed at

28.724 J/g, respectively, in agreement with the measured results shown in Table 2, which are 26.72 J/g and 34.17 J/g, respectively. Based on the calculated values shown in Table 2, (1) the specific heat sharply increases from 9.8 J/g to 26.72 J/g when the water concentration is increased from 4.5 % to 5.3 %; (2) the increase becomes moderate with higher water concentration; and (3) it decreases when the water concentration is higher than 8.6 %. In addition to that, the freezing peak temperature is also decreased. All these occur to coincide with the structure transition with increasing water

> **Water % Fractions Heat of Fusion (J/g) Peak T (oC)**  10.3 28.97 –18.88 8.6 34.17 –23.97 7.8 31.27 –22.74 5.3 26.72 –20.32 4.5 9.808 –20.63 3.6 2.48 -45 1.8 2.18 0

In addition to that, a total maximum heat capacity increase of 88 % is obtained in the 8.6 % water/PAO nanoemulsion fluids: for a temperature increase from −20 to 0 oC, 1 ml PAO absorbs 37.6 joules heat, and for the nanoemulsion containing 8.6 % water nanodroplets, the melting of ice nanoparticle absorbs 34 joules heat, it means that upon the

To gain further insight into that, the specific heat of each sample is also calculated and summarized here in Table 2. The calculated *H*<sup>f</sup> values of water-in-PAO nanoemulsions for different water concentrations from 5.3 to 8.6 vol. % are 17.7 and 28.724 J/g, respectively, in agreement with the measured results shown in Table 2, which are 26.72 J/g and 34.17 J/g, respectively. Based on the calculated values shown in Table 2, (1) the specific heat sharply increases from 9.8 J/g to 26.72 J/g when the water concentration is increased from 4.5 % to 5.3 vol. %; (2) the increase becomes moderate with higher water concentration; and (3) it decreases when the water concentration is higher than 8.6 vol. %. In addition to that, the freezing peak temperature is also decreased. All these occur to coincide with the structure transition with increasing water loading based on the SANS measurement.


**Table 2.** Specific heat of water-in-PAO nanoemulsion fluids

In addition to that, a total maximum heat capacity increase of 88 % is obtained in the 8.6 vol. % water/PAO nanoemulsion fluids: for a temperature increase from −20 to 0 o C, 1 ml PAO absorbs 37.6 joules heat, and for the nanoemulsion containing 8.6 % water nanodroplets, the melting of ice nanoparticle absorbs 34 joules heat, it means that upon the melting of the ice nanoparticles in the nanoemulsion, the heat capacity of PAO has increased up by about 76 % in addition to the enhancement in heat capacity caused only by the addition of water without phase change (~12%).

### **4. Conclusions**

The use of adding another material into thermal fluids has been emerged in recent years as a way to enhance the heat capacity and thermal conductivity of the base fluids simultaneously. A new type of nanostructured heat transfer fluids: nanoemulsion fluids are discussed in this chapter, such as ethanol-in-PAO, water-in-FC-72, and water-in-PAO nanoemulsion fluids. Many interesting properties have been reported in these nanoemulsion heat transfer fluids recently. The nanoemulsion heat transfer fluids can be formed by self-assembly and are thermodynamically stable. The self-assembled nanostructures have a significant effect on its macroscopic thermophysical properties which coincides with the structural characteristics measured using SANS. In addition to that, the effective heat capacity of base fluid can also be greatly enhanced when those phase changeable nanodroplets undergo phase transition: the effective heat capacity of FC-72 by more than 200 % when those droplets undergo liquid–solid phase transition in water-in-FC-72 nanoemulsion fluids, and the effective heat capacity of PAO is increased by 80 % in water-in-PAO nanoemulsion fluids. The use of nanoemulsion fluids provides a means to increase the fluid conductivity and heat capacity simultaneously in the base fluids and their application in a wide variety of applications appears promising, but several critical issues remain to be solved in the future, for example, large subcooling or superheating during phase change due to lack of nucleation sites, and large viscosity increase due to the dispersed nanodroplets inside.

### **Acknowledgements**

To gain further insight into that, the specific heat of each sample is also calculated and

different water concentrations from 5.3 to 8.6 vol. % are 17.7 and 28.724 J/g, respectively, in agreement with the measured results shown in Table 2, which are 26.72 J/g and 34.17 J/g, respectively. Based on the calculated values shown in Table 2, (1) the specific heat sharply increases from 9.8 J/g to 26.72 J/g when the water concentration is increased from 4.5 % to 5.3 vol. %; (2) the increase becomes moderate with higher water concentration; and (3) it decreases when the water concentration is higher than 8.6 vol. %. In addition to that, the freezing peak temperature is also decreased. All these occur to coincide with the structure transition with

**Water % Fractions Heat of Fusion (J/g) Peak T (oC)** 10.3 28.97 –18.88 8.6 34.17 –23.97 7.8 31.27 –22.74 5.3 26.72 –20.32 4.5 9.808 –20.63 3.6 2.48 -45 1.8 2.18 0

In addition to that, a total maximum heat capacity increase of 88 % is obtained in the 8.6 vol.

absorbs 37.6 joules heat, and for the nanoemulsion containing 8.6 % water nanodroplets, the melting of ice nanoparticle absorbs 34 joules heat, it means that upon the melting of the ice nanoparticles in the nanoemulsion, the heat capacity of PAO has increased up by about 76 % in addition to the enhancement in heat capacity caused only by the addition of water without

The use of adding another material into thermal fluids has been emerged in recent years as a way to enhance the heat capacity and thermal conductivity of the base fluids simultaneously. A new type of nanostructured heat transfer fluids: nanoemulsion fluids are discussed in this chapter, such as ethanol-in-PAO, water-in-FC-72, and water-in-PAO nanoemulsion fluids. Many interesting properties have been reported in these nanoemulsion heat transfer fluids recently. The nanoemulsion heat transfer fluids can be formed by self-assembly and are thermodynamically stable. The self-assembled nanostructures have a significant effect on its macroscopic thermophysical properties which coincides with the structural characteristics

% water/PAO nanoemulsion fluids: for a temperature increase from −20 to 0 o

values of water-in-PAO nanoemulsions for

C, 1 ml PAO

summarized here in Table 2. The calculated *H*<sup>f</sup>

98 Neutron Scattering

**Table 2.** Specific heat of water-in-PAO nanoemulsion fluids

phase change (~12%).

**4. Conclusions**

increasing water loading based on the SANS measurement.

The authors would like to thank Dr. Boualem Hammouda at Center for Neutron Research (National Institute of Standards and Technology, Gaithersburg, MD) for helping to conduct the SANS experiments and having a constructive discussion on SANS data post processing.

This study is financially supported by National Science Foundation (CBET-0730963). The SANS measurements performed at the NIST-CNR are supported in part by the National Science Foundation under Agreement No. DMR-0944772.

The identification of commercial products does not imply endorsement by the National Institute of Standards and Technology nor does it imply that these are the best for the purpose.

### **Author details**

Bao Yang2 and Jiajun Xu1\*

\*Address all correspondence to: jiajun.xu@udc.edu

1 Department of Mechanical Engineering, University of the District of Columbia, Washington DC, USA

2 Department of Mechanical Engineering, University of Maryland, College Park, MD, USA

### **References**

[1] Eastman, J. A., Phillpot, S. R., Choi, S. U. S., and Keblinski, P., 2004, "Thermal trans‐ port in nanofluids," Annual Review of Materials Research, 34, pp. 219–246.


[12] Prasher, R., Song, D., Wang, J. L., and Phelan, P., 2006, "Measurements of nanofluid viscosity and its implications for thermal applications," Applied Physics Letters, 89(13).

[2] Eastman, L. J., Choi, S. U. S., LI, S., and Thompson, L. J., 1997, "Enhanced thermal conductivity through development of nanofluids," Nanocrystalline and Nanocompo‐

[3] Inaba, H., 2000, "New challenge in advanced thermal energy transportation using functionally thermal fluids," International Journal of Thermal Sciences, 39(9–11), pp.

[4] Buongiorno, J., Venerus, D. C., Prabhat, N., McKrell, T., Townsend, J., Christianson, R., Tolmachev, Y. V., Keblinski, P., Hu, L.-W., Alvarado, J. L., Bang, I. C., Bishnoi, S. W., Bonetti, M., Botz, F., Cecere, A., Chang, Y., Chen, G., Chen, H., Chung, S. J., Chyu, M. K., Das, S. K., Di Paola, R., Ding, Y., Dubois, F., Dzido, G., Eapen, J., Escher, W., Funfschilling, D., Galand, Q., Gao, J., Gharagozloo, P. E., Goodson, K. E., Gutier‐ rez, J. G., Hong, H., Horton, M., Hwang, K. S., Iorio, C. S., Jang, S. P., Jarzebski, A. B., Jiang, Y., Jin, L., Kabelac, S., Kamath, A., Kedzierski, M. A., Kieng, L. G., Kim, C., Kim, J.-H., Kim, S., Lee, S. H., Leong, K. C., Manna, I., Michel, B., Ni, R., Patel, H. E., Philip, J., Poulikakos, D., Reynaud, C., Savino, R., Singh, P. K., Song, P., Sundarara‐ jan, T., Timofeeva, E., Tritcak, T., Turanov, A. N., Van Vaerenbergh, S., Wen, D., Witharana, S., Yang, C., Yeh, W.-H., Zhao, X.-Z., and Zhou, S.-Q., 2009, "A bench‐ mark study on the thermal conductivity of nanofluids," Journal of Applied Physics,

[5] Zimparov, V., 2002, "Energy conservation through heat transfer enhancement techni‐

[6] Dewan, A., Mahanta, P., Raju, K. S., and Kumar, P. S., 2004, "Review of passive heat transfer augmentation techniques," Proceedings of the Institution of Mechanical En‐

[7] Boyd, R. D., 1985, "Subcooled flow boiling critical heat-flux (CHF) and its application to fusion energy components. 1. A review of fudamentals of CHF and related data-

[8] Choi, S. U. S., Zhang, Z. G., Yu, W., Lockwood, F. E., and Grulke, E. A., 2001, "Anom‐ alous thermal conductivity enhancement in nanotube suspensions," Applied Physics

[9] Maxwell, J. C., 1904, "A treatise on electricity and magnetism," Oxford University

[10] Eastman, J. A., Choi, S. U. S., Li, S., Yu, W., and Thompson, L. J., 2001, "Anomalously increased effective thermal conductivities of ethylene glycol-based nanofluids con‐

[11] Keblinski, P., Phillpot, S. R., Choi, S. U. S., and Eastman, J. A., 2002, "Mechanisms of heat flow in suspensions of nano-sized particles (nanofluids)," International Journal

taining copper nanoparticles," Applied Physics Letters, 78(6), pp. 718–720.

ques," International Journal of Energy Research, 26(7), pp. 675–696.

gineers Part A-Journal of Power and Energy, 218(A7), pp. 509–527.

base," Fusion Technology, 7(1), pp. 7–30.

of Heat and Mass Transfer, 45(4), pp. 855–863.

Letters, 79(14), pp. 2252–2254.

Press, Cambridge, UK.

site Materials II.

991–1003.

100 Neutron Scattering

106(9).


[39] Ray, S., Bisal, S. R., and Moulik, S. P., 1994, "Thermodynamics of microemulsion for‐ mation. 2. Enthalpy of solution of water in binary-mixtures of aerosol-OT and hep‐ tane and heat-capacity of the resulting systems," Langmuir, 10(8), pp. 2507–2510.

[26] Bulanov, N. V., Skripov, V. P., and Khmylnin, V. A., 1993, "Heat transfer to tmulsion

[27] Bulanov, N. V., 2001, "An analysis of the heat flux density under conditions of boil‐

[28] Bulanov, N. V., and Gasanov, B. M., 2005, "Experimental setup for studying the chain activation of low-temperature boiling sites in superheated liquid droplets," Colloid

[29] Bulanov, N. V., Gasanov, B. M., and Turchaninova, E. A., 2006, "Results of experi‐ mental investigation of heat transfer with emulsions with low-boiling disperse

[30] Bulanov, N. V., and Gasanov, B. M., 2008, "Peculiarities of boiling of emulsions with a low-boiling disperse phase," International Journal of Heat and Mass Transfer, 51(7–

[31] Lunde, D. M., 2011, "Boiling dilute emulsions on a heated strip," MS thesis, Universi‐

[32] Chen, S. J., Evans, D. F., and Ninham, B. W., 1984, "Properties and structure of 3-com‐ ponent ionic microemulsions," Journal of Physical Chemistry, 88(8), pp. 1631–1634.

[33] Ruckenstein, E., 1986, "The surface of tension, the natural radius, and the interfacialtension in the thermodynamics of microemulsions," Journal of Colloid and Interface

[34] Siano, D. B., Bock, J., Myer, P., and Russel, W. B., 1987, "Thermodynamics and hydro‐ dynamics of a nonionic microemulsion," Colloids and Surfaces, 26, pp. 171–190. [35] Rosano, H. L., Cavallo, J. L., Chang, D. L., and Whittam, J. H., 1988, "Microemulsions - A commentary on their preparation," Journal of the Society of Cosmetic Chemists,

[36] Chen, Z. Q., Chen, L. D., Hao, C., and Zhang, C. Z., 1990, "Thermodynamics of mi‐ croemulsion. 1. The effect of alkyl chain-length of alkyl aromatics," Acta Chimica Sin‐

[37] Moulik, S. P., Das, M. L., Bhattacharya, P. K., and Das, A. R., 1992, "Thermodynamics of microemulsion formation. 1. Enthalpy of solution of water in binary (triton-X 100 + butanol) and ternary (heptane + triton-X 100 + butanol) mixtures and heat-capacity of

[38] Moulik, S. P., and Ray, S., 1994, "Thermodynamics of clustering of droplets in water/AOT/heptane microemulsion," Pure and Applied Chemistry, 66(3), pp. 521–

the resulting systems," Langmuir, 8(9), pp. 2135–2139.

with a low-boiling disperse phase," Heat Transfer Research, pp. 786–789.

ing internal phase of emulsion," High Temperature, 39(3), pp. 462–469.

Journal, 67(5), pp. 531–536.

8), pp. 1628–1632.

102 Neutron Scattering

ty of Minnesota.

39(3), pp. 201–209.

ica, 48(6), pp. 528–533.

525.

Science, 114(1), pp. 173–179.

phase," High Temperature, 44(2), pp. 267–282.


[51] Hammouda, B., Krueger, S., and Glinka, C. J., 1993, "Small-angle neutron-scattering at the national-institute-of-standards-and-technology," Journal of Research of the Na‐

[52] Chen, S. H., 1986, "Small-angle neutron-scattering studies of the structure and inter‐ action in micellar and microemulsion systems," Annual Review of Physical Chemis‐

[53] Gradzielski, M., and Langevin, D., 1996, "Small-angle neutron scattering experiments on microemulsion droplets: Relation to the bending elasticity of the amphiphilic

[54] Hammouda, B., 2010, "SANS from polymers-review of the recent literature," Polymer

[55] Howe, A. M., Toprakcioglu, C., Dore, J. C., and Robinson, B. H., 1986, "Small-angle neutron-scattering studies of microemulsions stabilized by aerosel-OT. 3. The effect of addictives on phase-stability and droplet structure," Journal of the Chemical Soci‐

[56] Marszalek, J., Pojman, J. A., and Page, K. A., 2008, "Neutron scattering study of the structural change induced by photopolymerization of AOT/D(2)O/sodecyl acrylate

[57] Nagao, M., Seto, H., Shibayama, M., and Yamada, N. L., 2003, "Small-angle neutron scattering study of droplet density dependence of the water-in-oil droplet structure in a ternary microemulsion," Journal of Applied Crystallography, 36, pp. 602–606. [58] Wang, X.-Q., and Mujumdar, A. S., 2007, "Heat transfer characteristics of nanofluids:

[59] Wang, L. Q., and Wei, X. H., 2009, "Nanofluids: synthesis, heat conduction, and ex‐

[60] Ozerinc, S., Kakac, S., and Yazicioglu, A. G., 2010, "Enhanced thermal conductivity of nanofluids: a state-of-the-art review," Microfluidics and Nanofluidics, 8(2), pp. 145–

[61] Philip, J., and Shima, P. D., 2012, "Thermal properties of nanofluids," Advances in

[62] Cahill, D. G., 1990, "Thermal-conductivity measurement from 30-K to 750-K - The 3-

[63] Yang, B., 2008, "Thermal conductivity equations based on Brownian motion in sus‐ pensions of nanoparticles (nanofluids)," Journal of Heat Transfer-Transactions of the

tional Institute of Standards and Technology, 98(1), pp. 31–46.

film," Journal of Molecular Structure, 383(1–3), pp. 145–156.

inverse microemulsions," Langmuir, 24(23), pp. 13694–13700.

a review," International Journal of Thermal Sciences, 46(1), pp. 1–19.

tension," Journal of Heat Transfer-Transactions of the Asme, 131(3).

omega method," Review of Scientific Instruments, 61(2), pp. 802–808.

ety-Faraday Transactions I, 82, pp. 2411–2422.

Colloid and Interface Science, 183, pp. 30–45.

try, 37, pp. 351–399.

104 Neutron Scattering

Reviews, 50(1), pp. 14–39.

170.

Asme, 130(4).


### **Application of Neutron Scattering in Amino Acid Crystals – Structural and Dynamical Information**

Paulo T.C. Freire, José E.M. Pereira and Heloisa N. Bordallo

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/62198

### **Abstract**

In this work results of neutron powder diffraction and inelastic neutron scattering on ali‐ phatic amino acid (L-alanine, L-valine and L-leucine) crystals are presented. The study al‐ lows inferring important informations about both the stability of the structures and the dynamics of phase transitions. In particular, it was possible to compare the behavior of different amino acids and to infer that changes in the mean-square displacements ob‐ served for two deuterated forms of L-alanine ND3 + – C2H4 – CO2 and NH3 + – C2D4 – CO2 are due to a structural rearrangement related to thermally activated amplitude reorientation of NH3 + and CO2 - . Conformation changes in L-valine and L-leucine related to the phase transition are also discussed.

**Keywords:** Amino acid crystal, neutron diffraction, inelastic neutron scattering

### **1. Introduction**

Amino acids are the basic blocks of proteins of all living beings, but are also found alone in the cytoplasm of several cells. They were produced and selected since the beginning of life in the Earth, although it is not possible to rule out an extraterrestrial origin. Due to the fact that they appear forming zwitterion structure (involving simultaneously the NH3 + and the CO2 groups) in both aqueous environment and in solid state phase, amino acids have been extensively studied in the last years. The researches deal mainly with spectroscopic and structural properties, giving special attention to the hydrogen bonds between the amino acid and the water molecules in the aqueous solution and among amino acids, when the material is found in crystalline structures [1 – 4].

From the spectroscopic point of view, amino acids have been investigated using mainly Raman spectroscopy [4 – 6], where information about optical phonons is readily obtained. Based on

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

such studies it has been suggested that amino acid crystals can present structural instabilities at low temperatures or high pressures. Regarding their structural properties, X-rays [7, 8] and neutron scattering diffraction data [5] have been useful in providing a picture of the origin of the stabilities of their crystal structures as well as the role played by the hydrogen bonds.

In particular, it was possible to gain information on the modifications of the hydrogen bonds studying different deuterated of amino acids samples using inelastic neutron scattering [9 – 11]. From these data the elastic intensity can be expressed as a function of temperature such as *I(T)* = *I(0).exp[-Q2 u(T)2 ]*, where *Q* is the scattering vector and *u(T)* represents the mean-square displacement of the scattering nuclei. Additionally, if a dynamic transition occurs, one can observe a change in the slope of the observed *u(T)2* . Furthermore, selective deuteration allows tackling changes in the mean-square displacements of particular groups, such as the torsion vibration of CH3 + or the torsion of NH3 + group. Inelastic neutron scattering combined to selective deuteration can also be used to highlight the modes of particular molecular groups by comparing the spectra of the selected deuterated groups [12].

The objective of this chapter is to present a quick review of neutron powder diffraction and inelastic neutron scattering results on various aliphatic amino acid crystals. These investiga‐ tions allowed to compare the behavior of different aliphatic amino acids (alanine, valine and leucine), in particular, looking at their vibrational and dynamical properties. For L-alanine, for example, combining neutron diffraction to incoherent elastic neutron scattering measurements allowed us to infer that changes in the mean-square displacements observed for two deuter‐ ated forms of L-alanine: (i) ND3 + – C2H4 – CO2 and (ii) NH3 + – C2D4 – CO2 are due to a structural rearrangement related to thermally activated amplitude reorientation of NH3 + and CO2 - .

### **2. Structural and vibrational anomalies observed in selected amino acids**

Many results on the vibrational properties of amino acid crystals are known. From these studies one has constructed an interesting picture about the behavior of the crystals in a huge range of temperature and/or pressure. L-alanine, the simplest chiral amino acid, is an example where surprising aspects of the vibrational properties stand. Among these aspects it is possible to cite some: the intriguing step-wise evolution of the wavenumber of some bands [13], the unconventional increasing of the c lattice parameter when L-alanine is cooled [14], and the localization of vibrational states as revealed by the variation of intensities of the low wave‐ number modes at 41 and 49 cm-1 [15]. Simultaneously to the increasing of the c lattice parameter a progressive conformational change of the NH3 + zwitterion, as well as a strong dynamic Jahn-Teller-like effect due the coupling between NH3 + charge and the lattice were observed [13]. These results can be interpreted as consequence of the N – H stretching self-trapped states related to the N – H...O hydrogen bonds directed along the head-to-tail chains of the zwitter‐ ions. According to Ref. [16] these self-trapped states should be associated with some of the anomalies reported for L-alanine. The behavior of the intensity of one lattice mode was realized as consequence of the response of one N – H...O hydrogen bond to respond to the changes in the interaction of the nitrogen and oxygen atoms with other atoms forming different hydrogen bonds. In fact, it is believed that one can assume that hydrogen bonds in chains of zwitterion have diverse properties from those between the chains and the variation of temperature should produce subtle rearrangement of the hydrogen bond network, explaining the existence of anomalies at low temperature [16]. Under high pressure, L-alanine also shows intriguing aspects. For instance, Raman spectroscopy studies have indicated the occurrence of a phase transition at ~ 2.3 GPa [17], which was later confirmed by X-ray diffraction measurements [18]. A new interpretation, however, has indicated that the modifications about 2 GPa might also be understood as a conformation change of the NH3 + group [19, 20].

such studies it has been suggested that amino acid crystals can present structural instabilities at low temperatures or high pressures. Regarding their structural properties, X-rays [7, 8] and neutron scattering diffraction data [5] have been useful in providing a picture of the origin of the stabilities of their crystal structures as well as the role played by the hydrogen bonds.

In particular, it was possible to gain information on the modifications of the hydrogen bonds studying different deuterated of amino acids samples using inelastic neutron scattering [9 – 11]. From these data the elastic intensity can be expressed as a function of temperature such

displacement of the scattering nuclei. Additionally, if a dynamic transition occurs, one can

tackling changes in the mean-square displacements of particular groups, such as the torsion

selective deuteration can also be used to highlight the modes of particular molecular groups

The objective of this chapter is to present a quick review of neutron powder diffraction and inelastic neutron scattering results on various aliphatic amino acid crystals. These investiga‐ tions allowed to compare the behavior of different aliphatic amino acids (alanine, valine and leucine), in particular, looking at their vibrational and dynamical properties. For L-alanine, for example, combining neutron diffraction to incoherent elastic neutron scattering measurements allowed us to infer that changes in the mean-square displacements observed for two deuter‐


**2. Structural and vibrational anomalies observed in selected amino acids**

Many results on the vibrational properties of amino acid crystals are known. From these studies one has constructed an interesting picture about the behavior of the crystals in a huge range of temperature and/or pressure. L-alanine, the simplest chiral amino acid, is an example where surprising aspects of the vibrational properties stand. Among these aspects it is possible to cite some: the intriguing step-wise evolution of the wavenumber of some bands [13], the unconventional increasing of the c lattice parameter when L-alanine is cooled [14], and the localization of vibrational states as revealed by the variation of intensities of the low wave‐ number modes at 41 and 49 cm-1 [15]. Simultaneously to the increasing of the c lattice parameter

+

These results can be interpreted as consequence of the N – H stretching self-trapped states related to the N – H...O hydrogen bonds directed along the head-to-tail chains of the zwitter‐ ions. According to Ref. [16] these self-trapped states should be associated with some of the anomalies reported for L-alanine. The behavior of the intensity of one lattice mode was realized as consequence of the response of one N – H...O hydrogen bond to respond to the changes in the interaction of the nitrogen and oxygen atoms with other atoms forming different hydrogen

+

and (ii) NH3

+

– C2D4 – CO2


zwitterion, as well as a strong dynamic Jahn-

charge and the lattice were observed [13].

are due to a structural

and CO2


+

+

*]*, where *Q* is the scattering vector and *u(T)* represents the mean-square

. Furthermore, selective deuteration allows

group. Inelastic neutron scattering combined to

as *I(T)* = *I(0).exp[-Q2 u(T)2*

+

ated forms of L-alanine: (i) ND3

vibration of CH3

108 Neutron Scattering

observe a change in the slope of the observed *u(T)2*

or the torsion of NH3

by comparing the spectra of the selected deuterated groups [12].

+

a progressive conformational change of the NH3

Teller-like effect due the coupling between NH3

– C2H4 – CO2

rearrangement related to thermally activated amplitude reorientation of NH3

The above considerations for the chiral L-alanine can be extended to encompass the racemic material. DL-alanine seems to be stable at low temperature, presenting no evidence of a solidsolid phase transition from room temperature down to 15 K [21]. However, although no modification in the spectral region of the lattice modes – normally between 30 and 200 cm-1 – was detected, certain changes in the internal mode region of the spectrum were noted. These changes can be understood as slight conformational modification of the molecule on cooling. In fact, the rocking vibration of CO2 presents an unconventional behavior, *i.e*., variation of the wavenumber in step-like shape at specific temperatures. In relation to this, it is important to mention that the data obtained using polarized Raman shows that the main change associated with hydrogen bonds take place along the x-axis direction, instead of the bonds along the zaxis. This behavior was also simultaneously reported in the work presented in Ref. [16] for the chiral partner of alanine crystal.

Crystalline L-leucine, another aliphatic amino acid, presents an interesting behavior when submitted to temperature variation between 300 and 400 K. As shown by Raman spectroscopic measurements, the material undergoes a phase transition at ~ 353 K. One possible explanation for the occurrence of such a modification is the appearance of transverse optical (TO) modes at high temperature. Such a vibrational aspect can be translated to the structural language if one assumes that the room temperature monoclinic structure C2 change to a monoclinic Cs structure [22]. In other words, from the point of view of the spectroscopic scrutiny the C2 → Cs phase transition is a viable path.

Obviously, other questions are relevant as the interaction of amino acid molecules with metal ions. Ions can be present in the living beings as participating of biochemistry processes or nosy hazard agents. Additionally, from a technological point of view it is important to know the influence of metal doping on several crystals with the objective to modify some of their properties, including modification of the crystal habit, thermal conductivity, etc. Analysis of Raman spectroscopy and X-ray diffraction measurements suggested, for example, Ni ions occupy interstitial crystallographic sites in the lattice of L-histidine hydrochloride monohy‐ drate crystal [23]. This specific position is the same occupied by Cu ions in the L-histi‐ dine.HCl.H2O crystal; the stabilization of the ion – amino acid molecule could be achieved by the interaction between the metal ion and the NH3 <sup>+</sup> group. On cooling, due to the packing of structure, the interaction ion – amino acid is intensified. To understand the interaction of metal ion and amino acid (representing active part of proteins and peptides) is a task with relevance to biochemistry science.

From this résumé one realizes that many changes might occur when crystalline amino acids are submitted to different pressure and/or temperature conditions. Therefore, one question comes to mind: how the hydrogen bonds connecting the molecules in the solid state phase adapt themselves to the new phases in these amino acids? This question will be addressed in the next sections, where it is explained how neutron diffraction and inelastic neutron scattering experiments can bring unique information to this problem.

### **3. Basic theory**

Postulated by Rutherford in 1920 and discovered by J. Chadwick in 1932, neutrons are noncharged subatomic particles with mass rather close to that of the proton *mn* = 1.675.10-27 kg, magnetic moment equals to - 1.913 µb and a nuclear spin *s* = ½, which are important and unique properties for developing the theory of neutron scattering [24].

Since neutrons are non-charged particles, their interactions with matter, both nuclear and magnetic, are short range; therefore neutrons can penetrate deeply into matter. This enables us to study the structure and dynamics of materials under very precise environmental conditions. Neutrons can be described using either classical mechanics, with momentum *p* <sup>→</sup> =*mv* <sup>→</sup> , where *m* is the neutron mass and *v* → is its velocity, or quantum mechanics, with mo‐ mentum *p* <sup>→</sup> =ℏ*k* <sup>→</sup> , with |*k* <sup>→</sup> <sup>|</sup> <sup>=</sup> <sup>2</sup>*<sup>π</sup> <sup>λ</sup>* defining the wave vector of the neutron and λ its associated wavelength.

This also implies that neutron scattering processes are based on the analysis of momentum and energy transfer, which may occur due to the interactions between neutrons and matter. The momentum transfer vector or the scattering vector, *Q* <sup>→</sup>, is defined as the vector difference between the incoming and scattered wave vectors as in Eq. (1).

$$
\overrightarrow{Q} = \overrightarrow{k\_i} - \overrightarrow{k\_{f'}} \tag{1}
$$

while the variation in the neutron momentum is given by Eq. (2):

$$
\hbar \vec{\mathbf{Q}} = \hbar (\overrightarrow{k\_i} - \overrightarrow{k\_f}) \,, \tag{2}
$$

The corresponding neutron energy *E* can be described as described by Eq. (3):

$$E = \frac{p^2}{2m} = \frac{1}{2}m\nu^2 = \frac{\hbar^2}{2m\lambda} = \frac{\hbar^2 k^2}{2m},\tag{3}$$

where *h* =2*π*ℏ = 6.626.10-34J.s is the Planck's constant. Thus, variation of the energy (ℏ*ω*) measured in a neutron scattering experiment is given by Eq. (4):

Application of Neutron Scattering in Amino Acid Crystals – Structural and Dynamical Information http://dx.doi.org/10.5772/62198 111

$$
\hbar \rho \alpha = E\_i - E\_f = \frac{\hbar^2}{2m} \left( k\_i^2 - k\_f^2 \right). \tag{4}$$

During the scattering process, elastic scattering occurs when the variation of energy between neutron and the sample is zero. On the other hand, when this variation is different from zero, the scattering processes is called inelastic.

From this résumé one realizes that many changes might occur when crystalline amino acids are submitted to different pressure and/or temperature conditions. Therefore, one question comes to mind: how the hydrogen bonds connecting the molecules in the solid state phase adapt themselves to the new phases in these amino acids? This question will be addressed in the next sections, where it is explained how neutron diffraction and inelastic neutron scattering

Postulated by Rutherford in 1920 and discovered by J. Chadwick in 1932, neutrons are noncharged subatomic particles with mass rather close to that of the proton *mn* = 1.675.10-27 kg, magnetic moment equals to - 1.913 µb and a nuclear spin *s* = ½, which are important and unique

Since neutrons are non-charged particles, their interactions with matter, both nuclear and magnetic, are short range; therefore neutrons can penetrate deeply into matter. This enables us to study the structure and dynamics of materials under very precise environmental conditions. Neutrons can be described using either classical mechanics, with momentum

This also implies that neutron scattering processes are based on the analysis of momentum and energy transfer, which may occur due to the interactions between neutrons and matter.

> , *Qk k i f* = r ur uur

( ), *Q kk i f* = -

<sup>2</sup> 2 22 <sup>1</sup> <sup>2</sup> , 22 2 2 *<sup>p</sup> <sup>k</sup> E m m mm* n

where *h* =2*π*ℏ = 6.626.10-34J.s is the Planck's constant. Thus, variation of the energy (ℏ*ω*)

l

The corresponding neutron energy *E* can be described as described by Eq. (3):

→ is its velocity, or quantum mechanics, with mo‐

<sup>→</sup>, is defined as the vector difference

(1)

*<sup>λ</sup>* defining the wave vector of the neutron and λ its associated

<sup>r</sup> ur uur h h (2)

== = = h h (3)

experiments can bring unique information to this problem.

properties for developing the theory of neutron scattering [24].

<sup>→</sup> , where *m* is the neutron mass and *v*

<sup>→</sup> <sup>|</sup> <sup>=</sup> <sup>2</sup>*<sup>π</sup>*

The momentum transfer vector or the scattering vector, *Q*

between the incoming and scattered wave vectors as in Eq. (1).

while the variation in the neutron momentum is given by Eq. (2):

measured in a neutron scattering experiment is given by Eq. (4):

<sup>→</sup> , with |*k*

**3. Basic theory**

110 Neutron Scattering

*p* <sup>→</sup> =*mv*

mentum *p*

wavelength.

<sup>→</sup> =ℏ*k*

In a typical neutron experiment the collected information is related to the partial differential cross-section, *d2 σ/*(*dEdΩ)*, given by Eq. (5), according to Ref. [25]:

$$\frac{d^2 \sigma}{d\Omega d E\_f} = \frac{k\_f}{k\_l} \frac{1}{\hbar} \left[ b\_{\rm coh}^2 S\_{\rm coh} \left( \vec{Q}\_{\prime} \, o \right) + b\_{\rm inc}^2 S\_{\rm inc} \left( \vec{Q}\_{\prime} \, o \right) \right],\tag{5}$$

where *Scoh* (*Q* <sup>→</sup>, *<sup>ω</sup>*) and *Sinc*(*<sup>Q</sup>* <sup>→</sup>, *<sup>ω</sup>*) are the so-called coherent and incoherent scattering functions. These functions are Fourier transformations of the van Hove correlation functions *G*(*r* <sup>→</sup> , *t*), which can be described as the probability for an atom at the origin at time 0, to also be found within the unit volume at a position *r* <sup>→</sup> at a time *t*.

The probability of interactions between the neutrons and the atoms is a characteristic of each isotope and described by two terms: coherent and incoherent cross sections (*σcoh* and *σinc*), as shown in Table 1. The cross section is related to the coherent (*bcoh*) and incoherent (*binc*) scattering length by the following relation: *σ = 4πb<sup>2</sup>* . The high *σinc* value of 1 H in comparison to the other elements allows the assignment of the vibrations related to hydrogen atoms. On the other, the high value of the coherent cross section of 2 D when compared to 1 H allows to locate the hydrogen atom by using deuteration.


**Table 1.** Coherent (*σcoh*) and incoherent (*σinc*) cross-sections in 10-24 cm2 for selected elements and isotopes [26].

For crystalline matrices, the coherent scattering function in Eq. (5) is written as given in Eq. (6):

$$\begin{split} \mathcal{S}\_{\text{out}} &= \left| F(\vec{Q}) \right|^2 \delta(o) \sum\_{\substack{\vec{C} \\ \frac{\vec{C}}{2\pi}}} \delta\left( \vec{Q} - \vec{G} \right) \\ &+ \sum\_{s,\vec{q}} \left| F\left(s, \vec{q}\right) \right|^2 \frac{\left\langle n(o) \right| + \frac{1}{2} \pm \frac{1}{2} \right\rangle}{o} \delta(o \mp o\_{s,\vec{q}}) \sum\_{\substack{\vec{C} \\ \frac{\vec{C}}{2\pi}}} \delta\left( \vec{Q} + \vec{q} - \vec{G} \right), \end{split} \tag{6}$$

where the first term represents Bragg scattering and in a diffraction experiment the Bragg's Law is satisfied:

$$
\hbar \mathbf{u} \mathcal{X} = \mathcal{Z}d \sin \theta\_{\prime} \tag{7}
$$

here *n* is a positive integer and λ is the wavelength of incident wave and the angle *θ* is defined as the angle between the incoming ray and the diffraction atomic plane.

The second term in Eq. (6) describes collective atomic motions (phonons), where *q* <sup>→</sup> is the wave vector of the collective atomic motion and *s* an index that specifies the various modes of such collective motions.

Contributions from single particle motion are described by *Sinc*(*Q* <sup>→</sup>, *<sup>ω</sup>*) as given by Eq. (8):

$$\mathcal{S}\_{\rm inc} \left( \vec{\mathbf{Q}}\_{\prime} \, \boldsymbol{\alpha} \right) = \mathbf{S}\_{\rm inc}^{\rm trans} \left( \vec{\mathbf{Q}}\_{\prime} \, \boldsymbol{\alpha} \right) \otimes \mathbf{S}\_{\rm inc}^{\rm rot} \left( \vec{\mathbf{Q}}\_{\prime} \, \boldsymbol{\alpha} \right) \otimes \mathbf{S}\_{\rm inc}^{\rm vib} \left( \vec{\mathbf{Q}}\_{\prime} \, \boldsymbol{\alpha} \right) \, \tag{8}$$

In Eq. (8) the terms describe contributions from the translational, rotational and vibrational modes, respectively.

### **4. Experimental details**

In order to obtain the results presented in this chapter, a series of facilities were utilized. Neutron powder diffraction (NPD) experiments were carried out using the powder diffrac‐ tometers E9 located at the Helmholtz-Zentrum Berlin für Materialien und Energie (Berlin, Germany) and D2b at the Institut Laue-Langevin (Grenoble, France). Such equipments allowed the determination and refinement of crystal structures with very high resolution, furnishing lattice constants and internal atomic coordinates with great precision. Through the experi‐ ments with temperature variation the data were analyzed using a crystallographic model from Ref. [27] Due to the thermal diffuse scattering contribution, attention was paid to the modeling of the background.

Incoherent elastic neutron scattering measurements were carried out using the time-of-flight spectrometer NEAT at the Helmholtz-Zentrum Berlin für Materialien und Energie as well as the backscattering instrument IN10 located at the Institut Laue-Langevin.

The analysis of the evolution of the elastic line centered around *E* = 0, (energy resolution, *∆E* = 100 µeV (NEAT) or 1 µeV (IN10)) as a function of temperature gives information about the mean square displacement of the atoms around their equilibrium positions or, in very broad terms, describes the crystal flexibility. On the other hand, the inelastic part of the spectra gives information on the lattice phonons as well as on the inter-molecular vibrations of the samples. The obtained information is complementary to infrared (IR) and Raman scattering (RS).

### **5. Looking inside amino acid crystals**

where the first term represents Bragg scattering and in a diffraction experiment the Bragg's

 q

here *n* is a positive integer and λ is the wavelength of incident wave and the angle *θ* is defined

vector of the collective atomic motion and *s* an index that specifies the various modes of such

( , , , ,, ) ( ) ( ) ( ) *trans rot vib inc inc inc inc SQ S Q SQ SQ*

In Eq. (8) the terms describe contributions from the translational, rotational and vibrational

In order to obtain the results presented in this chapter, a series of facilities were utilized. Neutron powder diffraction (NPD) experiments were carried out using the powder diffrac‐ tometers E9 located at the Helmholtz-Zentrum Berlin für Materialien und Energie (Berlin, Germany) and D2b at the Institut Laue-Langevin (Grenoble, France). Such equipments allowed the determination and refinement of crystal structures with very high resolution, furnishing lattice constants and internal atomic coordinates with great precision. Through the experi‐ ments with temperature variation the data were analyzed using a crystallographic model from Ref. [27] Due to the thermal diffuse scattering contribution, attention was paid to the modeling

Incoherent elastic neutron scattering measurements were carried out using the time-of-flight spectrometer NEAT at the Helmholtz-Zentrum Berlin für Materialien und Energie as well as

The analysis of the evolution of the elastic line centered around *E* = 0, (energy resolution, *∆E* = 100 µeV (NEAT) or 1 µeV (IN10)) as a function of temperature gives information about the mean square displacement of the atoms around their equilibrium positions or, in very broad terms, describes the crystal flexibility. On the other hand, the inelastic part of the spectra gives information on the lattice phonons as well as on the inter-molecular vibrations of the samples. The obtained information is complementary to infrared (IR) and Raman scattering (RS).

the backscattering instrument IN10 located at the Institut Laue-Langevin.

www

= ÄÄ r rrr (8)

= 2 sin , (7)

<sup>→</sup> is the wave

<sup>→</sup>, *<sup>ω</sup>*) as given by Eq. (8):

*n d* l

The second term in Eq. (6) describes collective atomic motions (phonons), where *q*

as the angle between the incoming ray and the diffraction atomic plane.

Contributions from single particle motion are described by *Sinc*(*Q*

w

Law is satisfied:

112 Neutron Scattering

collective motions.

modes, respectively.

of the background.

**4. Experimental details**

In relation to amino acid crystals there is a pivotal point to be considered when using neutron spectroscopy or neutron diffraction: the hydrogen bonds that connect the various molecules together in the crystal structure. In this chapter results on crystals of the amino acids L-alanine, L-valine and L-leucine are discussed.

The first discussion will furnish a picture on the simplest chiral amino acid, L-alanine. Selective deuteration of L-alanine were analyzed: (i) C2H4-(ND3 + )-CO2 - , Ala-ND3; (ii) C2D4-(NH3 + )-CO2 - , Ala-CD4; (iii) C2D4-(ND3 + )-CO2 - , Ala-D7. L-alanine crystallizes in an orthorhombic structure with four molecules per unit cell in a *P212121* space group. The molecules are distributed spatially as head-to-tail chains along the crystallographic *c*-axis, presenting a complex three dimensional network of hydrogen bonds. Here we recall that the *c*-axis a stepwise change of the size parameter on cooling. Combining inelastic neutron scattering (INS) and neutron powder diffraction (NPD) is possible to study the relationship between the local dynamics of NH3, CO2 and CH3 and to gain insights about the behavior of hydrogen bonds.

Figure 1 presents the inelastic neutron scattering spectra of (a) Ala-ND3 and (b) Ala-CD4 for several temperatures in the energy range from 0 to 70 meV. Data were obtained using the NEAT spectrometer. It is possible to observe several normal modes: lattice modes below 20 meV and other modes such as torsion of CO2 group, τ(CO<sup>2</sup> - ), torsion of NH3, τ(NH3 + ), and torsion of CH3, τ(CH3), as specified in the figure. It is interesting to observe the peak located ~ 60 meV that is associated with τ(NH<sup>3</sup> + ). The width of this peak can be understood as conse‐ quence of the anharmonicity of the hydrogen bonds because all three protons participate of N–H...O hydrogen bonds. The influence of anharmonicity is not restrict to the torsion of the ammonia group; as it can be noted in Figure 1 the large peaks associated with the lattice modes that involves bending and stretching of hydrogen bonds can be ascribed to the same effect. Additionally, if one plots the energy of τ(CO<sup>2</sup> - ) mode in Ala-ND3 (not shown in Figure 1) as a function of temperature, one notes a discontinuity between 150 and 200 K. Such a discontinuity coincides with a discontinuity of the wavenumber of a band associated with τ(CO<sup>2</sup> - ) in the Raman spectrum at temperatures in the same interval.

Figure 2 presents the neutron diffraction patterns of Ala-CD4 for two different temperatures. The measurements were obtained with a neutron wavelength of 2.8060 Å. Impressively, at T = 100 K, it is observed a Bragg peak at ~ 2θ = 78° that does not belong to the *P212121* orthorhombic structure. In the attempt to index this peak authors were guided to assume a primitive orthorhombic cell with the *a*-axis with double size [11]. It is remembered that the nondeuterated L-alanine presents any strong modification in its lattice parameter, although the crystal itself shows a series of anomalies at low temperatures. As a consequence, the result presented in Figure 2 suggests that the partial deuteration of L-alanine induces a solid-solid phase transition. Interestingly such a phase transition can be seen as an enhancement of an anomaly observed in the non-deuterated partner at T ~ 260 K [13]. The doubling of the *a*-lattice parameter involves changes in the two hydrogen bonds represented in Figure 3 and a rearrangement of the whole molecule. Between 270 and 250 K, a decreasing in the D(2)...O(2)

**Figure 1.** Inelastic neutron scattering spectra at various temperatures for (a) Ala-ND3 and (b) Ala-CD4. The assignment of the main modes is given [11].

bond is observed and an impressive anomaly between 250 and 100 K. On the other hand, the D(1)...O(1) bond becomes longer when temperature is lower than 100 K.

Figure 4 presents <*u*H(*T*)2 > of C2H4-(ND3 + )-CO2 and C2D4-(NH3 + )-CO2 - , where *u*H(*T*) represents the mean-square displacement of the scattering nuclei. Here, as specified previously, *I(T)* = *I(0).exp*[*-Q2 u(T)2* ] is the elastic intensity expressed as a function of temperature (the tempera‐ ture dependence is embedded in *u*H). In fact, once the elastic intensity has been experimentally determined, one obtain *u(T)2* . As it is expected, due the large incoherent cross section of the hydrogen, *u(T)* is supposed originated only the motion of H atoms. Another aspect to be

Application of Neutron Scattering in Amino Acid Crystals – Structural and Dynamical Information http://dx.doi.org/10.5772/62198 115

**Figure 2.** Neutron diffraction patterns of C2D4-(NH3 + )-CO2 at T=300 K and T=100 K measured with the E9 diffractometer at Hahn-Meitner-Institut. The arrow indicates the peak that could not be indexed with the original *P212121* structure [11].

bond is observed and an impressive anomaly between 250 and 100 K. On the other hand, the

**Figure 1.** Inelastic neutron scattering spectra at various temperatures for (a) Ala-ND3 and (b) Ala-CD4. The assignment

the mean-square displacement of the scattering nuclei. Here, as specified previously, *I(T)* =

ture dependence is embedded in *u*H). In fact, once the elastic intensity has been experimentally

hydrogen, *u(T)* is supposed originated only the motion of H atoms. Another aspect to be

and C2D4-(NH3

] is the elastic intensity expressed as a function of temperature (the tempera‐

+ )-CO2 -

. As it is expected, due the large incoherent cross section of the

, where *u*H(*T*) represents

D(1)...O(1) bond becomes longer when temperature is lower than 100 K.

+ )-CO2 -

> of C2H4-(ND3

Figure 4 presents <*u*H(*T*)2

of the main modes is given [11].

determined, one obtain *u(T)2*

*I(0).exp*[*-Q2 u(T)2*

114 Neutron Scattering

considered is the following: change in the slope of *u*(*T*) 2 indicates a dynamical transition in the material. From the Figure 4 one observe that at 160 K and at 220 K there are clear changes in the slop, that can be interpreted as a phase transition or, at least, to a structural rearrangement. Such rearrangement should be probably related to thermally activated large-amplitude reorientations of both CH3 and NH3 smithereens of the L-alanine molecule.

The previous paragraphs showed that in the study of partially deuterated L-alanine crystal, the H(1)...O(1) hydrogen bond length increases below 100 K on decreasing temperature and

**Figure 3.** Representation of the crystal structure of L-alanine showing the *ab*-plane. The two hydrogen bonds modifies at low temperature are shown [11].

**Figure 4.** Mean-square dependence from the normalized elastic intensity as a function of temperature. The left scale and open circles represent data for C2H4-(ND3 + )-CO2 - and the right scale and full circles represent data for C2D4-(NH3 + )- CO2 - [10].

the H(2)...O(2) length presents an anomaly between 100 and 250 K. What about the behavior of fully deuterated L-alanine under the scrutiny of INS and NPD techniques? Roughly speaking, deuteration, or isotopic substitution, D → H, produces strong modification in the length of the hydrogen bond. This change generally affects the tunneling rates and the motional properties of the molecules, an effect known as Ubbelohde effect; it is observed in crystals with complex network of hydrogen bonds and even in isolated pair of this type of bond.

Figure 5 presents the temperature dependence of several parameters related to the hydrogen bonds for the fully deuterated L-alanine crystal, C2D4-(ND<sup>3</sup> + )-CO<sup>2</sup> - , Ala-D7. Several anomalies are described in the next paragraphs.

From the measurement it is possible to realize that the N-D(3) distance remains almost constant between 10 and 270 K. On the other hand, the N-D(1) distance remains constant only between 10 and 130 K, and diminishes for temperatures higher than 130 K (Figure 5a). Also interesting is the dramatic increase of the D(3)...O(2) hydrogen bond responsible for connecting the molecules into columns (see Figure 5b). This last point is in complete accordance with the fact that torsion of NH3 group presents important anharmonicity, as depicted during the discussion of Figure 1. So, one can understand the anharmonicity of τ(NH<sup>3</sup> + ) as consequence of the modification of one specific hydrogen bond during the temperature variation [10]. Looking at the average C–D lengths one notes anomalies at low temperatures as presented in Figure 5c. From the Figure 5d one also notes an anomaly in the C–C–C bond angle that extends from about 160 K up to 225 K. In this point is important to emphasize that Raman spectroscopy studies showed splitting of bands associated with lattice modes at T ~ 170 K. This means evidence of a phase transition. Additionally, it was observed anomalies in the Debye-Waller factor, confirming the phase transition at about 170 K in the fully deuterated L-alanine [10].

L-valine was the second aliphatic amino acid crystal studied through neutron techniques. Investigations so performed revealed aspects related to hydrogen bonds, ultimately, respon‐ sible for the structural stability of the material. L-valine crystallizes in a monoclinic structure with *P21* space group and four molecules per unit cell. From these molecules, two assumes *gauche I* conformation and two others *trans* conformation. Previous work using Raman spectroscopy suggested L-valine should undergo a solid-solid phase transition because impressive modifications of the bands associated with lattice modes were observed at low temperatures [28].

Figure 6 presents the dynamical susceptibility (χ´´(E)) of both L-valine hydrogenated [(CH3)2CHCH(NH2)CO2H] and deuterated [(CD3)2CDCD(NH2)CO2H] (synthesized by Dr. Ehrenstorfer GmbH from Augsburg, Germany). The comparison of the INS spectra of L-valine and deuterated L-valine (L-valine-d8) shows differences that can be understood as follows. At first, an intense band at 480 cm-1 in the spectrum of L-valine-d8 points to the accuracy of the assignment of the vibrations involving the N–H...O hydrogen bonds. Also, a decrease of the signal for energy lower than 450 cm-1 indicates they are associated with the CH3 and CO<sup>2</sup> groups and the N–C–C units [9]. It is impressive in the room temperature INS spectrum of L-valine the presence of a mode at ~ 120 cm-1. It is remembered that such a band not visible at T= 300 K appeared in the Raman spectra of L-valine for temperatures lower than 120 K. In other words, this band could be associated with a mode not active in the Raman spectrum of the monoclinic room temperature structure, being active only in the infrared. During the phase transition the mode should become Raman active and the phase transition should be realized as consequence

the H(2)...O(2) length presents an anomaly between 100 and 250 K. What about the behavior of fully deuterated L-alanine under the scrutiny of INS and NPD techniques? Roughly

**Figure 4.** Mean-square dependence from the normalized elastic intensity as a function of temperature. The left scale


+ )-

+ )-CO2

**Figure 3.** Representation of the crystal structure of L-alanine showing the *ab*-plane. The two hydrogen bonds modifies

at low temperature are shown [11].

116 Neutron Scattering

and open circles represent data for C2H4-(ND3

CO2 - [10].

**Figure 5.** Temperature dependence of (a) N–D, (b) D...O, (c) average C–D length, (d) avergage C–C–C bond angles for fully deuterated L-alanine. The molecule is represented in the top left side of the figure an dashed lines are guides to the eyes [10].

of the activation of an infrared mode. An additional aspect of the study is to show (Figure 6b) the splitting of the band at ~ 15 meV at low temperature when compared with the spectrum recorded at 300 K, reinforcing the idea of a phase transition, as supported by the Raman spectroscopic study [28].

Up to now the present authors have seen several aspects of the bonds on L-alanine and an explanation for the phase transition undergone by L-valine using neutrons. It can be noted

Application of Neutron Scattering in Amino Acid Crystals – Structural and Dynamical Information http://dx.doi.org/10.5772/62198 119

**Figure 6.** (a) Dynamical susceptibility [χ´´(E)] for both L-valine (full squares) and L-valine-d8 (open squares) using a 5.1 Å incident wavelength at the NEAT spectrometer. (b) χ´´(E) of L-valine for two temperatures, 300 K and 120 K. The arrow in part (a) corresponds to the mode at 120 cm-1, observed in the low temperature range by Raman spectroscopy [12].

of the activation of an infrared mode. An additional aspect of the study is to show (Figure 6b) the splitting of the band at ~ 15 meV at low temperature when compared with the spectrum recorded at 300 K, reinforcing the idea of a phase transition, as supported by the Raman

**Figure 5.** Temperature dependence of (a) N–D, (b) D...O, (c) average C–D length, (d) avergage C–C–C bond angles for fully deuterated L-alanine. The molecule is represented in the top left side of the figure an dashed lines are guides to

Up to now the present authors have seen several aspects of the bonds on L-alanine and an explanation for the phase transition undergone by L-valine using neutrons. It can be noted

spectroscopic study [28].

the eyes [10].

118 Neutron Scattering

that subtle conformational change can lead to impressive modifications in the structural and vibrational properties of these two members of the amino acid family. Now, the discussion deals about another aliphatic amino acid. L-leucine, NH3-CH-CO2-CH2-CH-(CH3)2, crystallizes in a monoclinic structure *P21* with four molecules per unit cell. The carboxyl and amino groups are hydrogen bonded in a double layer, forming the crystal structure. The van der Waals interaction plays its role by connecting neighboring layers [1]. Due the existence of an addi‐ tional CH2 group in its chain, L-leucine is slightly more hydrophobic than L-valine. A previous research demonstrated the occurrence of a phase transition in L-leucine at ~ 353 K possibly maintaining the crystal with a monoclinic structure [22].

Figure 7 presents the dynamical susceptibility function χ´´ obtained from calculations based on the inelastic data from NEAT. It is perceived that below 280 K the mode at 30 meV (ap‐ proximately at 240 cm-1) presents a red shift down to 190 K and below this, presents a blue shift; also, the intensity increases on cooling. Interesting enough is the occurrence of a redistribution of intensity of the bands between 300 and 375 K that can be correlated with the phase transition observed at 353 K. The phase transition can also be glimpsed by the intensities of most modes located below 30 meV, associated with lattice modes, that show pronounced changes in the interval 300 – 375 K.

**Figure 7.** (a) Dynamical susceptibility [χ´´(E)] for L-leucine in the temperature range from 130 and 375 K measured in the energy interval from 1 to 30 meV. The mode at ~ 30 meV is associated with an out-of-plane vibration of CH2 [29].

Supposing S(Q,ω) the scattering function, with Q being the magnitude of the scattering wave vector and ω being the energy transfer, it is possible to decompose it in three different components: SE(Q, ω = 0), the elastic; SE(Q, ω ~ 0), quasi-elastic; SE(Q, ω > 0), inelastic. The study of the elastic scattering decay can furnish information about transitions because it appears as changes in the temperature dependence of the elastic intensity. Figure 8a shows the elastically scattered intensity as a function of temperature for L-leucine. The main contribution to SE comes from the hydrogen atoms. Within the time scale of ~ 10 ps originated from the resolution of the equipment one can argue that the transition at ~ 150 K is a consequence of CH3 groups. Such an anomaly confirms differential scanning calorimetry (DSC), as well as X-ray diffraction measurements [29].

In addition, one notes the broadening of the quasi-elastic response apparently disappears in the temperature range 220 – 250 K and the peak height decreases. It was possible to derive a

Application of Neutron Scattering in Amino Acid Crystals – Structural and Dynamical Information http://dx.doi.org/10.5772/62198 121

tional CH2 group in its chain, L-leucine is slightly more hydrophobic than L-valine. A previous research demonstrated the occurrence of a phase transition in L-leucine at ~ 353 K possibly

Figure 7 presents the dynamical susceptibility function χ´´ obtained from calculations based on the inelastic data from NEAT. It is perceived that below 280 K the mode at 30 meV (ap‐ proximately at 240 cm-1) presents a red shift down to 190 K and below this, presents a blue shift; also, the intensity increases on cooling. Interesting enough is the occurrence of a redistribution of intensity of the bands between 300 and 375 K that can be correlated with the phase transition observed at 353 K. The phase transition can also be glimpsed by the intensities of most modes located below 30 meV, associated with lattice modes, that show pronounced

**Figure 7.** (a) Dynamical susceptibility [χ´´(E)] for L-leucine in the temperature range from 130 and 375 K measured in the energy interval from 1 to 30 meV. The mode at ~ 30 meV is associated with an out-of-plane vibration of CH2 [29].

Supposing S(Q,ω) the scattering function, with Q being the magnitude of the scattering wave vector and ω being the energy transfer, it is possible to decompose it in three different components: SE(Q, ω = 0), the elastic; SE(Q, ω ~ 0), quasi-elastic; SE(Q, ω > 0), inelastic. The study of the elastic scattering decay can furnish information about transitions because it appears as changes in the temperature dependence of the elastic intensity. Figure 8a shows the elastically scattered intensity as a function of temperature for L-leucine. The main contribution to SE comes from the hydrogen atoms. Within the time scale of ~ 10 ps originated from the resolution of the equipment one can argue that the transition at ~ 150 K is a consequence of CH3 groups. Such an anomaly confirms differential scanning calorimetry (DSC), as well as X-ray diffraction

In addition, one notes the broadening of the quasi-elastic response apparently disappears in the temperature range 220 – 250 K and the peak height decreases. It was possible to derive a

maintaining the crystal with a monoclinic structure [22].

changes in the interval 300 – 375 K.

120 Neutron Scattering

measurements [29].

**Figure 8.** (a) Elastically scattered intensity as a function of temperature for L-leucine. An anomaly in the temperature evolution is marked by an arrow. (b) Dynamic structure factors obtained for AE = 98 µeV for several temperatures averaged over the whole range of Q. (c) Lorentzian half width at half maximum (HWHM), G, obtained using an Ar‐ renhius-like relation [29].

qualitative description by fitting the quasi-elastic SQE, using a Dirac function and a Lorentzian linewidth, which could describe, respectively, the elastic and quasi-elastic signals. In this approach it is possible to correlate the quasi-elastic broadening (Γ) with the activation energy, Eact, by the relation Γ = Γ0.exp[-Eact/kT]. The most important point derived from this relation is presented in Figure 8c, where a discontinuity, associated with the high temperature phase transition is observed between 345 and 360 K.

### **6. Conclusions**

Neutron diffraction and inelastic neutron scattering are powerful techniques to investigate several materials, among them, amino acid crystals, as it was shown in this chapter. From these studies it was presented the possibilty to correlate small variations of the network of hydrogen bonds to conformation changes in the amino acid molecules.

L-leucine, that presents only one polymorph at room temperature, was shown to undergo phase transitions at low and high temperatures. The new phases were analyzed and insights about the conformation of the molecules were attained.

In the case of L-valine it was possible through comparison between Raman spectroscopy and inelastic neutron scattering to identify infrared active modes responsible for the phase transition undergone by the crystal at approximately 120 K.

Finally, a series of information was obtained for the simplest proteic amino acid crystal, Lalanine. In the case of the fully deuterated sample, L-alanine-d7, our results showed a clear evidence of a structural phase transition at ~ 170 K. Such fact can be related to the Ubbe‐ lohde effect, or in other words different dimensions of the hydrogen bonds stabilize differently the deuterated and the non-deuterated crystals. Consequently, although the deuterated sister presents a phase transition, the non-deuterated structure, even if a series of structural and vibrational anomalies are observed, is stable at low temperatures. Our investigations point to the fact that the Ubbelohde effect can be significant, even if only small changes of hydrogen bond dimensions occur.

### **Acknowledgements**

PTCF acknowledges financial support from CNPq and FUNCAP agencies. JEMP work is financed by the Brazilian "Science Without Borders" Program.

### **Author details**

Paulo T.C. Freire1 , José E.M. Pereira2 and Heloisa N. Bordallo2

\*Address all correspondence to: tarso@fisica.ufc.br

1 Departamento de Física – Universidade Federal do Ceará, Fortaleza, Brazil

2 Niels Bohr Institute – University of Copenhagen, Copenhagen, Denmark

### **References**

**6. Conclusions**

122 Neutron Scattering

Neutron diffraction and inelastic neutron scattering are powerful techniques to investigate several materials, among them, amino acid crystals, as it was shown in this chapter. From these studies it was presented the possibilty to correlate small variations of the network of

L-leucine, that presents only one polymorph at room temperature, was shown to undergo phase transitions at low and high temperatures. The new phases were analyzed and insights

In the case of L-valine it was possible through comparison between Raman spectroscopy and inelastic neutron scattering to identify infrared active modes responsible for the phase

Finally, a series of information was obtained for the simplest proteic amino acid crystal, Lalanine. In the case of the fully deuterated sample, L-alanine-d7, our results showed a clear evidence of a structural phase transition at ~ 170 K. Such fact can be related to the Ubbe‐ lohde effect, or in other words different dimensions of the hydrogen bonds stabilize differently the deuterated and the non-deuterated crystals. Consequently, although the deuterated sister presents a phase transition, the non-deuterated structure, even if a series of structural and vibrational anomalies are observed, is stable at low temperatures. Our investigations point to the fact that the Ubbelohde effect can be significant, even if only small

PTCF acknowledges financial support from CNPq and FUNCAP agencies. JEMP work is

and Heloisa N. Bordallo2

hydrogen bonds to conformation changes in the amino acid molecules.

about the conformation of the molecules were attained.

transition undergone by the crystal at approximately 120 K.

financed by the Brazilian "Science Without Borders" Program.

1 Departamento de Física – Universidade Federal do Ceará, Fortaleza, Brazil

2 Niels Bohr Institute – University of Copenhagen, Copenhagen, Denmark

, José E.M. Pereira2

\*Address all correspondence to: tarso@fisica.ufc.br

changes of hydrogen bond dimensions occur.

**Acknowledgements**

**Author details**

Paulo T.C. Freire1


solid-solid phase transition. ChemPhysChem. 2009; 10 3337- 3343. DOI: 10.1002/cphc. 200900482.


the structure of L-histidine hydrochloride monohydrate, crystals, Journal of Physics: Condensed Matter. 2008; 20: 275209. DOI: 10.1088/0953-8984/20/27/275209.

[24] Lovesey SW. Neutron Scattering Theory. Oxford: Oxford Science Publications; 1986. 317 p. DOI: 10.1088/0953-8984/5/34/016.

solid-solid phase transition. ChemPhysChem. 2009; 10 3337- 3343. DOI: 10.1002/cphc.

[12] Silva JH, Lima Jr, JA, Freire PTC, Lemos V, Mendes Filho J, Melo FEA, Pizani PS, Fischer J, Klemke B, Kemner E, Bordallo HN. Raman spectroscopy and inelastic neu‐ tron scattering study of crystalline L-valine., Journal of Physics: Condensed Matter.

[13] Barthes M, Bordallo HN, Dénoyer F, Lorenzo J-E, Zaccaro J, Robert A, Zontone F. Mi‐ cro-transitions or breathers in L-alanine? The European Physical Journal B. 2004; 37:

[14] Destro R, Marsh RE, Bianchi R. A low-temperature (23K) study of L-alanine. The Journal of Physical Chemistry. 1988; 92: 966-973 DOI: 10.1021/j100315a022.

[15] Migliori A, Maxton PM, Clogston AM, Zirngiebl E, Lowe M. Anomalous tempera‐ ture dependence in the Raman spectra of l-alanine: Evidence for dynamic localiza‐ tion. Physical Review B. 1988; 38: 13464-13467. DOI: 10.1103/PhysRevB.38.13464.

[16] Kolesov BA, Boldyreva EV. Micro-conformational transition in L-alanine single crys‐ tals revisited by low wavenumber Raman spectroscopy. Journal of Raman Spectro‐

[17] Teixeira AMR, Freire PTC, Moreno AJD, Sasaki JM, Ayala AP, Mendes Filho J, Melo FEA. High-pressure Raman study of l-alanine crystal. Solid State Communications.

[18] Olsen JS, Gerward L, Freire PTC, Mendes Filho J, Melo FEA, Souza Filho AG. Pres‐ sure-induced phase transitions in L-alanine crystals. Journal of Physics and Chemis‐

[19] Funnel NP, Dawson A, Francis D, Lennie DR, Marshall WG, Moggach SA, Warren JE, Parsons S. The effect of pressure on the crystal structure of l-alanine. CrystEng‐

[20] Tumanov NA, Boldyreva EV, Kolesov BA, Kurnosov AV, Quesada Cabrera R. Pres‐ sure-induced phase transitions in L-alanine, revisited. Acta Crystallographica Section

[21] Lima Jr. JÁ, Melo FEA, Mendes Filho J, De Sousa G.P, Lima RJC, Façanha Filho PF, Bordallo HN. Low-temperature Raman spectra of racemate DL-Alanine crystals.

[22] Façanha Filho PF, Freire PTC, Lima KCV, Mendes Filho J, Melo FEA, Pizani PS. High temperature Raman spectra of L-leucine crystals. Brazilian Journal of Physics. 2008;

[23] Remédios CMR, Paraguassu W, Lima Jr. JÁ, Freire PTC, Mendes Filho J, Melo FEA, Menezes AS, dos Santos AO, Cardoso LP, Miranda MAR. Effect of Ni(II) doping on

Journal of Raman Spectroscopy. 2010; 41: 808-813. DOI: 10.1002/jrs.2507.

2009; 21: 415404. DOI: 10.1088/0953-8984/21/41/415404.

375-382. DOI: 10.1140/epjb/e2004-00069-1.

scopy. 2010; 42: 696-705. DOI: 10.1002/JRS.2768.

2000; 116: 405-409. DOI: 10.1016/S0038-1098(00)00342-2.

Comm. 2010; 12: 2573-2583. DOI: 10.1039/C001296C.

B. 2010; 66: 358-371. DOI: 10.1107/s010876811001983X.

38: 131-137. DOI: 10.1590/S0103-97332008000100024.

try of Solids. 2008; 69: 1641-1645. DOI: 10.1016/j.jpcs.2007.12.005.

200900482.

124 Neutron Scattering


### **Neutron Reflectometry for Studying Proteins/Peptides in Biomimetic Membranes**

Joyee Chun In Yeung, Tsung-Wu Lin and Hsin-Hui Shen

Additional information is available at the end of the chapter

http://dx.doi.org/10.5772/62781

### **Abstract**

The development of biomimetic surfaces for protein and peptide adsorptions is continu‐ ously expanding. Their biological functions can be influenced by the properties of the un‐ derlying artificial environment but the detailed mechanism is still not clear. In the past 30 years, neutron reflectometry has been widely applied to characterise the molecular struc‐ ture of proteins or multi-protein complexes and their interactions with fluid artificial membrane that mimics the cellular environment. The specific interactions, bindings or structural changes between proteins and membranes play a crucial role in cellular re‐ sponses and have promising potential in diagnostics and other biosensor applications. This chapter presents the progression of surface design for protein adsorption/interac‐ tions on membranes in detail, ranging from a simple phospholipid monolayer setup to more complicated artificial lipid bilayer systems. Furthermore, a new development of de‐ signed surfaces for studying the integral membrane protein system is also discussed in this chapter. A brief overview of various membrane mimetic surfaces is first outlined, fol‐ lowed by presenting specific examples of protein-membrane interactions studied by neu‐ tron reflectometry. The author demonstrates how to use neutron reflectometry as an advanced technique to provide step-by-step structural details for biomolecular applica‐ tions in a well-controlled manner.

**Keywords:** Neutron reflectometry, biomimetic surfaces, artificial membrane, proteins, peptides

### **1. Introduction**

Proteins are vital molecules that are responsible for many critical biological processes such as photosynthesis, respiration, cellular signal transduction, molecular transportation and enzymatic catalysis [1]. Most of these biochemical processes involve the interactions with proteins that are embedded in or associated with lipid molecules [2]. The protein-membrane

© 2016 The Author(s). Licensee InTech. This chapter is distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/3.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

interactions are dynamic, complicated and mainly driven by hydrophobic and electrostatic interactions, but the detailed mechanism is still not clear [3]. Understanding and further modulating the interactions between proteins and membranes are essential for many biological and biotechnological applications. Many *in vitro* studies have been conducted to reveal and monitor the molecular interactions between proteins and lipids [3, 4]. The proteins are adsorbed or immobilised on a biomimetic surface that reconstitutes the physiological envi‐ ronment, i.e., the use of artificial membrane models to mimic the biological membrane [5]. Neutron reflectometry is a powerful and non-destructive technique that has been widely employed for studying biomolecular interactions. It has unique properties to provide quanti‐ tative structural and compositional details of the model interfaces without physical damage to the sample, which is impossible to achieve using other techniques [6, 7].

Neutron reflectometry is a powerful tool to probe the interfacial chemical structures at the microscopic level. The setup for neutron reflectometry measurements requires a parallel neutron beam that is incident onto the sample deposited onto an optically flat surface, where the neutron beams reflect and exit from the opposite end [6]. The neutron reflectivity, R, refers to the ratio of the incoming and exiting neutron beams and it is measured as a function of momentum transfer *Q*, which is defined as *Q* =4*π sinθ* / *λ*, where θ is the angle of incident and <sup>λ</sup> is the neutron wavelength [6, 8, 9]. *R*(*Q*) is approximately given by *R*(*Q*)= <sup>16</sup>*<sup>π</sup>* <sup>2</sup> *<sup>Q</sup>* <sup>2</sup> <sup>|</sup>*ρ*(*Q*)| <sup>2</sup> , where *R*(*Q*) is the one-dimensional Fourier transform of *ρ*( z), the scattering length density (SLD) distribution normal to the surface. SLD is the sum of the coherent neutron scattering lengths of all atoms in the sample layer divided by the volume of the layer. Therefore, the intensity of the reflected beams is highly dependent on the thickness, densities and interface roughness of the samples [4, 6]. If there are mixed layers containing two components, A and B and water (w), the scattering length density will be *ρ* =*ϕAρ<sup>A</sup>* + *ϕBρ<sup>B</sup>* + *ϕwρw*, where ϕ is the volume fraction of each layer and *ρ* can be different contrasts from the same layer/substance, called contrast variations. The use of contrast variations in neutron reflectometry experiments allows resolving molecular species in response to stimuli; hence, it is possible to obtain detailed structure information with very high resolution [9]. There are two different approaches for contrast variations: magnetic contrast and isotopic contrast. Polarised neutron can generate a magnetic contrast that is able to produce simultaneously two datasets from exactly the same layer without changing the chemical environment [10, 11]. On the other hand, the isotopic approach is often used in the characterisation of multilayer, biological films [7]. By changing different buffer compositions (D2O, H2O and D2O/H2O mixtures) or using selective deuteration to replace hydrogen in desire locations, the contrast of scattering density between the hydrogen and deuterium reveals the relative locations in multi-component systems [12, 13]. By moni‐ toring the isotopic labels, the distributions of different components in an *in-situ* environment can be identified; hence the accurate compositional depth profiles and thickness of the biological film can be determined [4, 14]. Typically, this method is applied to investigate the structural orientation and conformation changes of proteins adsorbed on membranes in a noncrystalline aqueous environment [12, 13]. The structural information of proteins/or peptides and membranes can be determined at Angstrom nanometre scale.

Researchers have developed different artificial membrane models to reconstitute the essential features of the biological membranes, providing an appropriate system to investigate the structure-function relationships of proteins in membrane-mediated processes such as proteinor peptide-membrane interactions [15, 16]. This review discusses and examines several protein adsorption studies on different biomimetic membrane models used in neutron reflectometry. The biomimetic membrane models used are phospholipid monolayer at air/water interface (Section 2), phospholipid monolayer on solid supports (Section 3) and fluid lipid bilayers on a solid support (Section 4). The preparation of each membrane model is briefly described and characterised by neutron reflectometry, with particular attention on their biological applica‐ tions in protein/or peptide adsorption. Furthermore, there are reviews that describe a direct immobilisation of membrane proteins on gold-modified surface (Section 5), as an alternative method for protein adsorption studies. This method is suitable for the use of various immo‐ bilising surfaces and addresses the problem of maintaining the proper orientation of the immobilised proteins, hence should have a wide range of *in vitro* applications. The summary of the key and significant findings is stated.

interactions are dynamic, complicated and mainly driven by hydrophobic and electrostatic interactions, but the detailed mechanism is still not clear [3]. Understanding and further modulating the interactions between proteins and membranes are essential for many biological and biotechnological applications. Many *in vitro* studies have been conducted to reveal and monitor the molecular interactions between proteins and lipids [3, 4]. The proteins are adsorbed or immobilised on a biomimetic surface that reconstitutes the physiological envi‐ ronment, i.e., the use of artificial membrane models to mimic the biological membrane [5]. Neutron reflectometry is a powerful and non-destructive technique that has been widely employed for studying biomolecular interactions. It has unique properties to provide quanti‐ tative structural and compositional details of the model interfaces without physical damage

Neutron reflectometry is a powerful tool to probe the interfacial chemical structures at the microscopic level. The setup for neutron reflectometry measurements requires a parallel neutron beam that is incident onto the sample deposited onto an optically flat surface, where the neutron beams reflect and exit from the opposite end [6]. The neutron reflectivity, R, refers to the ratio of the incoming and exiting neutron beams and it is measured as a function of momentum transfer *Q*, which is defined as *Q* =4*π sinθ* / *λ*, where θ is the angle of incident and

*R*(*Q*) is the one-dimensional Fourier transform of *ρ*( z), the scattering length density (SLD) distribution normal to the surface. SLD is the sum of the coherent neutron scattering lengths of all atoms in the sample layer divided by the volume of the layer. Therefore, the intensity of the reflected beams is highly dependent on the thickness, densities and interface roughness of the samples [4, 6]. If there are mixed layers containing two components, A and B and water (w), the scattering length density will be *ρ* =*ϕAρ<sup>A</sup>* + *ϕBρ<sup>B</sup>* + *ϕwρw*, where ϕ is the volume fraction of each layer and *ρ* can be different contrasts from the same layer/substance, called contrast variations. The use of contrast variations in neutron reflectometry experiments allows resolving molecular species in response to stimuli; hence, it is possible to obtain detailed structure information with very high resolution [9]. There are two different approaches for contrast variations: magnetic contrast and isotopic contrast. Polarised neutron can generate a magnetic contrast that is able to produce simultaneously two datasets from exactly the same layer without changing the chemical environment [10, 11]. On the other hand, the isotopic approach is often used in the characterisation of multilayer, biological films [7]. By changing different buffer compositions (D2O, H2O and D2O/H2O mixtures) or using selective deuteration to replace hydrogen in desire locations, the contrast of scattering density between the hydrogen and deuterium reveals the relative locations in multi-component systems [12, 13]. By moni‐ toring the isotopic labels, the distributions of different components in an *in-situ* environment can be identified; hence the accurate compositional depth profiles and thickness of the biological film can be determined [4, 14]. Typically, this method is applied to investigate the structural orientation and conformation changes of proteins adsorbed on membranes in a noncrystalline aqueous environment [12, 13]. The structural information of proteins/or peptides

Researchers have developed different artificial membrane models to reconstitute the essential features of the biological membranes, providing an appropriate system to investigate the

*<sup>Q</sup>* <sup>2</sup> <sup>|</sup>*ρ*(*Q*)| <sup>2</sup>

, where

to the sample, which is impossible to achieve using other techniques [6, 7].

128 Neutron Scattering

<sup>λ</sup> is the neutron wavelength [6, 8, 9]. *R*(*Q*) is approximately given by *R*(*Q*)= <sup>16</sup>*<sup>π</sup>* <sup>2</sup>

and membranes can be determined at Angstrom nanometre scale.

### **2. Biomolecules interact with phospholipid monolayer at air/water interface**

Phospholipid monolayer at the air/water interface has been used as the simplest biomimetic membrane model [17]. A lipid monolayer can be formed by depositing the lipid solution on the surface of water within a Langmuir-Blodgett trough. The amphiphilic properties of the lipid molecules allow the hydrophilic headgroups to interact with the water and leaving the hydrophobic alkyl chains exposed to the air, forming a planar, two-dimensional lipid phase. The resulting homogeneous lipid monolayer is called a Langmuir monolayer with high reproducibility [17, 18] (Figure 1). The lipid packing density of the phospholipid monolayer formedusingtheLangmuirtroughcanbepreciselymonitoredbycontrollingtheappliedsurface pressure (π) as a function of the molecular area (A) at a constant temperature. At a low surface pressure, the disordered phospholipid molecules exist in the gas phase. When a lateral surface force is applied, the phospholipid molecules are compressed into the liquid-expanded phase, followed by the coexistence of the liquid-expanded and the liquid-condensed phases. Further increaseinthesurfacepressureallowsthephospholipidmoleculestoorientinanorderlymanner and presents as the solid state. Once the applied pressure reaches a certain limit, the phospho‐ lipid monolayer collapses [17, 19]. The organisation and dynamics of the phospholipid monolayer at the air/water interface have received immense interests in their biological applications. This model setup allows us to tailor the lipid packing density, which has a direct influence on the protein or peptide interactions with membranes [18]. The ability to monitor the lipid packing density throughout the experiment is beneficial and unique about this biomimetic membrane, which is not easy to control in many artificial membrane models [18].

The physical properties of phospholipid monolayer at the air/water interface influenced protein adsorption, as demonstrated in Maierhofer and co-workers' neutron reflectometry study [20]. The zwitterionic phospholipid monolayer used in the study consisted of 1,2 dipalmitoyl-*sn*-glycero-3-phosphocholine (DPPC) mixed with various molar ratios of 1,2 dimyristoyl-*sn*-glycero-3-phospho-(1'-rac-glycerol) (DMPG), to evaluate the surface charges

**Figure 1.** Diagram of phospholipid monolayer deposited on air/water interface prepared in a Langmuir trough. The phospholipid hydrophilic head group interacts with the aqueous solution, leaving the phospholipid hydrophobic tails exposed to the air. Protein adsorption occurs from the aqueous phase to the phospholipid monolayer.

of phospholipid monolayer in affecting cytochrome c adsorption. Higher protein adsorption was observed on the DPPC monolayer mixed with a 30 molar percent of DMPG, for which the DPPC/DMPG monolayer, existed in liquid-condensed phase. However, similar amount of cytochrome c was adsorbed on the equal molar mixture of DPPC and DMPG monolayer, known as an ideal lipid mixing [20]. The result strongly suggested that the electrostatic interactions between cytochrome c and the DPPC/DMPG monolayers were driven by the domain formation during phase transition of the non-ideal lipid mixing of the phospholipid monolayer that is predominantly at the liquid-condensed phase rather than the liquidexpanded phase [20].

Miano et al. [21] observed a constant thickness of 16 Å when DPPC monolayer was deposited on the air/water interface by applying various surface pressure using Langmuir-Blodgett trough. The molecular density (area per molecule) varied from 40 to 70 Å2 accompanying in the change of surface pressure [21]. The neutron reflectometry study from Miano et al. also showed that the amount of protein associated with membranes decreased when the lipid monolayer surface pressure increased. Another study investigated the interactions of lacto‐ ferrin, a typical tear protein, on the DPPC monolayer deposited at the air/water interface, as a model of preocular tear film outer interface. The amount of lactoferrin coadsorbed onto the DPPC monolayer decreased as the surface pressure on the DPPC monolayer increased [21]. The coadsorbed lactoferrin penetrated into the phospholipid monolayer and characterised into two layers: 20% of the protein phase was located at a top dense sublayer (15 Å) and 10% of the protein phase was adsorbed to the bottom sublayer (60 Å) [21]. The structural distribution of the coadsorbed lactoferrin layers provided a mechanistic detail of how the tear proteins interact with the lipid film, as an *in vitro* preocular model for evaporative tear loss [21].

Recently, the use of Langmuir trough in combination with neutron reflectometry also illus‐ trated how packing density could be altered and resulted in different conformations of the proteins [18, 22]. Pirrone et al. [18] monitored the membrane association process of human immunodeficiency virus-1 Negative Regulatory Factor (HIV-1 Nef) protein on a phospholipid monolayer formed on air/water interface [18]. The conformation of HIV-1 Nef proteins was found to be sensitive and dependent on lipid packing density. At a high lipid packing density (35 mN/m), the HIV-1 Nef protein maintained in a compact conformation and did not insert into the phospholipid monolayer. In a low lipid packing density (20 mM/m), both n-terminal region and c-terminal unstructured loop of the HIV-1 Nef proteins underwent conformation changes [18, 22]. The repositioning of HIV-1 Nef proteins caused the hydrophobic residues on the N-terminal inserted into the membrane and the proteins were oriented away from the phospholipid monolayer to a position 70 Å away [18, 22].

### **3. Biomolecules interact with phospholipid monolayer on solid supports**

The amphiphilic nature of lipid molecules allows a spontaneous formation of Langmuir-Blodgett lipid film on a solid support, including glass, silica, micas and metal substrates [17]. However, the hydrophobic tail of lipid molecules does not encourage the formation of lipid monolayer on hydrophilic substrates [23, 24]. The technical challenge is overcome by chemi‐ cally grafting the solid support with a polymer or surfactant, for example, octadecyltrichlor‐ osilane. The surface hydroxyl groups of octadecyltrichlorosilane form covalent siloxy bonds with the substrate, resulting in a self-assembled layer and leaving the hydrophobic layer for subsequent phospholipid adsorption. The strong hydrophobic interactions between the lipid molecules and the robust hydrophobic part of surfactant allow for the phospholipids to be deposited using a Langmuir-Blodgett technique and hence promotes the formation of homogeneous monolayer (Figure 2) [23, 24].

of phospholipid monolayer in affecting cytochrome c adsorption. Higher protein adsorption was observed on the DPPC monolayer mixed with a 30 molar percent of DMPG, for which the DPPC/DMPG monolayer, existed in liquid-condensed phase. However, similar amount of cytochrome c was adsorbed on the equal molar mixture of DPPC and DMPG monolayer, known as an ideal lipid mixing [20]. The result strongly suggested that the electrostatic interactions between cytochrome c and the DPPC/DMPG monolayers were driven by the domain formation during phase transition of the non-ideal lipid mixing of the phospholipid monolayer that is predominantly at the liquid-condensed phase rather than the liquid-

**Figure 1.** Diagram of phospholipid monolayer deposited on air/water interface prepared in a Langmuir trough. The phospholipid hydrophilic head group interacts with the aqueous solution, leaving the phospholipid hydrophobic tails

exposed to the air. Protein adsorption occurs from the aqueous phase to the phospholipid monolayer.

Miano et al. [21] observed a constant thickness of 16 Å when DPPC monolayer was deposited on the air/water interface by applying various surface pressure using Langmuir-Blodgett trough. The molecular density (area per molecule) varied from 40 to 70 Å2 accompanying in the change of surface pressure [21]. The neutron reflectometry study from Miano et al. also showed that the amount of protein associated with membranes decreased when the lipid monolayer surface pressure increased. Another study investigated the interactions of lacto‐ ferrin, a typical tear protein, on the DPPC monolayer deposited at the air/water interface, as a model of preocular tear film outer interface. The amount of lactoferrin coadsorbed onto the DPPC monolayer decreased as the surface pressure on the DPPC monolayer increased [21]. The coadsorbed lactoferrin penetrated into the phospholipid monolayer and characterised into two layers: 20% of the protein phase was located at a top dense sublayer (15 Å) and 10% of the protein phase was adsorbed to the bottom sublayer (60 Å) [21]. The structural distribution of the coadsorbed lactoferrin layers provided a mechanistic detail of how the tear proteins interact

with the lipid film, as an *in vitro* preocular model for evaporative tear loss [21].

Recently, the use of Langmuir trough in combination with neutron reflectometry also illus‐ trated how packing density could be altered and resulted in different conformations of the proteins [18, 22]. Pirrone et al. [18] monitored the membrane association process of human immunodeficiency virus-1 Negative Regulatory Factor (HIV-1 Nef) protein on a phospholipid monolayer formed on air/water interface [18]. The conformation of HIV-1 Nef proteins was found to be sensitive and dependent on lipid packing density. At a high lipid packing density (35 mN/m), the HIV-1 Nef protein maintained in a compact conformation and did not insert into the phospholipid monolayer. In a low lipid packing density (20 mM/m), both n-terminal region and c-terminal unstructured loop of the HIV-1 Nef proteins underwent conformation changes [18, 22]. The repositioning of HIV-1 Nef proteins caused the hydrophobic residues on

expanded phase [20].

130 Neutron Scattering

**Figure 2.** Phospholipid monolayer deposited on chemically grafted solid support, the hydrophobic interactions forms the phospholipid monolayer onto the self-assembled hydrophobic octadecyltrichlorosilane-modified surface.

The surface coverage of octadecyltrichlorosilane formed on the solid support was found to be 68%, with the thickness of 24.0 Å to 33.5 Å [24, 25]. Subsequently, phospholipid monolayers were deposited on the octadecyltrichlorosilane-modified solid support and their structural information was extracted from neutron reflectometry (Figure 2). Kuhl et al. [23] claimed that the structure of phospholipid monolayers was not affected by the quality of octadecyltrichlor‐ osilane deposited underneath. The octadecyltrichlorosilane-lipid layer was further studied in detail by Hollinshead et al. [24]. They observed that the phospholipid molecules penetrated into the octadecyltrichlorosilane layer and reported as a four-layer model. The first layer was the solid support (22.6 Å); the second layer (17.9 Å) and the third layer (12.6 Å) consisted of a combination of octadecyltrichlorosilane and phospholipid hydrocarbon chains and the top layer (9.4 Å) contained the phospholipid headgroup molecules only [24]. On the other hand, Kuhl et al. also demonstrated that the phospholipid monolayer, 1,2-distearoyl-sn-glycero-3 phosphoethanolamine (DSPE), could be functionalised with polymer such as polyethylene glycol (PEG). The DSPE-PEG monolayer gave the thickness of 52 Å on quartz surface and 48 Å on silicon surface. Thickness fluctuations of the PEG layer were adjusted by varying the concentration of polyethylene glycol in the phospholipid monolayer [23].

Lu et al. [26] reported the formation of phospholipid monolayer, phosphorylcholine (PC), on octadecyltrichlorosilane had a thickness of 18 Å [26]. The use of phospholipid monolayer mimics one leaflet of a biological membrane, and therefore it is ideal to investigate the interactions of peripheral membrane proteins [18]. The membrane interactions between lysozyme and albumin have been extensively studied and well-characterised [26, 27]. These proteins were selected as the model proteins for undertaking the neutron reflectometry experiments to reveal their interfacial structural conformations when adsorbed onto the phospholipid monolayer. The adsorption of lysozyme and albumins on the phosphorylcholine monolayer formed a thick and diffuse layer with the thickness of 60 Å and 80 Å, suggesting a loose protein layer deposited on phosphorylcholine monolayer [26, 27].

The other application of phospholipid monolayer on a solid support is to investigate the sensitivity of biodiagnostic assays [26] showing non-specific interaction of protein molecules. To retain the protein bioactivity, the phospholipid monolayer can be functionalised with a ligand such as biotin that is widely used to control protein orientation during adsorption [28]. Biotin bound to the headgroup of the phospholipid molecules and formed a biotin-lipid layer, followed by the binding of tetrameric protein, streptavidin, on the biotin-lipid layer. Neutron reflectometry results from Schmidt et al. [28] showed that only 5% of the DPPC molecules were carrying a biotin moiety. The layer thickness of DPPC monolayer, biotin moiety and the streptavidin proteins were 24 Å, 10 Å and 41 Å, respectively [28], that can be well defined by neutron reflectometry.

### **4. Biomolecules associated with fluid lipid bilayers on a solid support**

The use of fluid lipid bilayer membrane as a model platform to study the influence of the bilayer microenvironment in protein-membrane interactions has been rapidly developed [15, 16]. Various lipid bilayer models exist, including (a) single supported lipid bilayers, (b) tethered lipid bilayer membranes, (c) polymer-cushioned bilayer membranes and (d) floating supported bilayers as described in Figure 3 [29].

**Figure 3.** Different types of supported lipid bilayers on a solid support: (a) single supported lipid bilayers, (b) tethered lipid bilayer membranes, (c) polymer-cushioned bilayer membranes and (d) floating supported bilayers.

### **4.1. Single supported lipid bilayers**

Single supported lipid bilayers (SLBs) are commonly employed as the simplest lipid bilayer model in which the lipid bilayers are directly deposited on a hydrophilic surface (Figure 3a) [29]. SLBs are the most popular artificial membrane model because of their long-term stability, ease of formation and manipulation and wide applicability in studying membrane-related features and applications [29]. The following three methods are often used individually or in combination to the formation of SLBs: lipid vesicle fusion, lipid-detergent method and Langmuir-Blodgett and Langmuir*-*Schaefer deposition [29].

### *4.1.1. Lipid vesicle fusion*

Å on silicon surface. Thickness fluctuations of the PEG layer were adjusted by varying the

Lu et al. [26] reported the formation of phospholipid monolayer, phosphorylcholine (PC), on octadecyltrichlorosilane had a thickness of 18 Å [26]. The use of phospholipid monolayer mimics one leaflet of a biological membrane, and therefore it is ideal to investigate the interactions of peripheral membrane proteins [18]. The membrane interactions between lysozyme and albumin have been extensively studied and well-characterised [26, 27]. These proteins were selected as the model proteins for undertaking the neutron reflectometry experiments to reveal their interfacial structural conformations when adsorbed onto the phospholipid monolayer. The adsorption of lysozyme and albumins on the phosphorylcholine monolayer formed a thick and diffuse layer with the thickness of 60 Å and 80 Å, suggesting a

The other application of phospholipid monolayer on a solid support is to investigate the sensitivity of biodiagnostic assays [26] showing non-specific interaction of protein molecules. To retain the protein bioactivity, the phospholipid monolayer can be functionalised with a ligand such as biotin that is widely used to control protein orientation during adsorption [28]. Biotin bound to the headgroup of the phospholipid molecules and formed a biotin-lipid layer, followed by the binding of tetrameric protein, streptavidin, on the biotin-lipid layer. Neutron reflectometry results from Schmidt et al. [28] showed that only 5% of the DPPC molecules were carrying a biotin moiety. The layer thickness of DPPC monolayer, biotin moiety and the streptavidin proteins were 24 Å, 10 Å and 41 Å, respectively [28], that can be well defined by

**4. Biomolecules associated with fluid lipid bilayers on a solid support**

The use of fluid lipid bilayer membrane as a model platform to study the influence of the bilayer microenvironment in protein-membrane interactions has been rapidly developed [15, 16]. Various lipid bilayer models exist, including (a) single supported lipid bilayers, (b) tethered lipid bilayer membranes, (c) polymer-cushioned bilayer membranes and (d) floating

**Figure 3.** Different types of supported lipid bilayers on a solid support: (a) single supported lipid bilayers, (b) tethered

lipid bilayer membranes, (c) polymer-cushioned bilayer membranes and (d) floating supported bilayers.

concentration of polyethylene glycol in the phospholipid monolayer [23].

loose protein layer deposited on phosphorylcholine monolayer [26, 27].

neutron reflectometry.

132 Neutron Scattering

supported bilayers as described in Figure 3 [29].

Lipid vesicle fusion is a spontaneous reaction, where the lipid vesicles adsorb, rupture and form a lipid bilayer which covers greater than 80% of a solid hydrophilic support [30]. This process is dependent on the surface properties (charge, structure and roughness), lipid nature (composition, charge, size and physical state) as well as the nature of the solvent (pH and ionic strength) [29, 31, 32]. The lipid vesicle fusion technique was used for studying the formation of SLBs made of different types of unsaturated phospholipids, including 1-palmitoyl-2-oleoylsn-glycero-3-phosphocholine (POPC), 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-rac-(1 glycerol)] (POPG), 1-palmitoyl-2-oleoyl-sn-glycero-3-[phospho-l-serine] (POPS), 1,2-dioleoyl*sn*-glycero-3-phosphocholine (DOPC) and 1,2-dioleoyl-*sn-*glycero-3-phospho-l-serine (DOPS) [33–35]. The experiments were set above the phospholipid phase transition temperature, thus the lipid bilayers existed in the fluid phase. The formation of SLBs is symmetric between the two lipid layers, and the membrane structure can be modelled from the neutron reflectometry data [33–35]. The thickness of each layer varied depending on the types of phospholipids used to form SLBs. The lipid bilayer had an overall thickness ranging between 38 ± 3Å and 56 ± 3Å, which were separated into three discrete layers: head group one (in contact with bulk solvent), acyl tails and head group two (in contact with the solid support). The modelling data also yielded information on other structural parameters including area per lipid molecule, volume fraction and surface excess of the three discrete layers (head-tail-head) [33–35].

The integrity of SLBs formed by lipid vesicle fusion technique has been fully characterised using neutron reflectometry, and it has been shown as a promising application to study membrane-mediated processes. Fernandez et al. [36] studied the interactions of antimicrobial peptide, maculatin 1.1, on both eukaryotic- and prokaryotic-mimic membranes using neutron reflectometry. The cationic maculatin 1.1 showed a minimal effect to the eukaryotic-mimic membrane; whereas, a profound effect was found when interacting with the anionic head group on the prokaryotic-mimic membrane. A decrease in anionic head group thickness was observed, but the thickness of acyl tails region remained the same, suggesting a reordering of bilayer occurred to associate with transmembrane pore formation [36]. Hellstrand et al. [34] also used neutron reflectometry to determine the position of α-synuclein, an amyloid protein associated with Parkinson's disease, within and outside of the membrane bilayer. The αsynuclein was found to embed in the bilayer outer head group via electrostatic interactions but did not penetrate into the hydrophobic acyl chain region. The study then compared the adsorption of α-synuclein on different biological relevant membranes which were composed of pure POPC, the mixtures of POPC/POPS or cardiolipin. The effect of membrane fluidity (acyl chain saturation), head group separation and the electrostatic shielding of the phospho‐ lipids strongly influenced the association of α-synuclein onto the membranes [34]. An analogue study was carried out on prion protein, which is widely recognised as the causative agent for Parkinson's disease when the protein is misfolded [33]. The interactions of N-terminal cleavage fragments (N1 and N2) of prion protein on negatively charged POPG-contained SLBs were probed using neutron reflectometry. The results implied that both N1 and N2 were inserted into phospholipid head groups; however, the interactions were stronger for N1 fragment because of the additional metal ion binding site located at a polybasic region. The insertion of N1 fragment induced an increase in lipid order in the absence of phase transition, as evidenced by lengthening of lipid acyl tails and decreasing in lipid area. It is possible that prion protein N-terminal fragment plays a functional role on membrane interactions [33]. Recently, Lu et al. [35] employed neutron reflectometry to define the molecular determinants of cellular protein layers adsorbed on mimetic apoptotic membranes. The interactions between annexin V and membranes contained different amount of phosphatidylserine (PS) phospholipids were examined, which is a crucial process to detect early/mid-stage cell membrane apoptosis *in vivo* and *in vitro*. The membrane properties that were affected by PS contents strongly affected annexin V binding, showing an increase in annexin V binding with an increase in PS content in the membrane and that there was a maximum binding at 20% of PS in the membrane. A decrease in the thickness of the adsorbed annexin V layer was observed with a 33% of PS content in membrane, suggesting the annexin V might alter its conformational change at high PS content [35].

### *4.1.2. Lipid-detergent method*

Lipid-detergent method is an alternative method for depositing SLBs on solid support. Tiberg et al. [37, 38] have successfully demonstrated the formation of SLB with the aid of detergent incorporated into the lipid forming micelles. The presence of detergents allows saturated fatty acids such as DPPC that has high hydrophobicity, to form stable micelles in aqueous solution [39]. In this method, the detergent is first incorporated into the lipid solution in excess amount, this helps to emulsify the lipid molecules and raise to the critical micelle concentration. Once lipid molecules concentration reached the critical micelle concentration, they aggregated into stable lipid micelles, which were introduced and adsorbed onto a hydrophilic silica surface [37–40]. The resulted SLB was identical to those formed using the vesicle fusion method once the detergent was completely removed [29, 38, 40, 41]. The cholesterol incorporated DPPC membranes were also used to study the effect of added components on the resulting SLBs. The neutron reflectometry results showed that the cholesterol was located below the lipid head group region and subsequently led to an increase in membrane thickness [38].

In addition to the membrane structural information, neutron reflectometry was also used to study protein interactions with membranes [29, 42]. Vacklin et al. [42] investigated the physical interactions between phospholipase A2 (PLA2) upon regulating its enzymatic hydrolysis on supported lipid bilayer. They were able to monitor the membrane bilayer changing its morphology when exposed to PLA2 and the reaction was irreversible. By using phospholipids with a different degree of saturation, the rate of PLA2 hydrolysis decreased as lipid saturation increased. This gave the smallest penetration depth on the unsaturated phospholipid, 1,2 dioleoyl-sn-glycero-3-phosphocholine (DOPC); and the greatest penetration depth on the saturated phospholipid, DPPC [42]. Shen et al. [43] studied the interactions of lipopeptide surfactin on supported DPPC bilayers, observing that the membrane stability was highly dependent on the critical micelle concentration of surfactin. The membrane was stable at a low surfactin concentration (below the critical micelle concentration). Neutron reflectometry can further identify the accumulation of surfactin on membranes showing that the surfactin penetrated into the outer leaflet of the membrane bilayer with pore formations on the mem‐ brane. When the surfactin concentration reached or exceeded the critical micelle concentration, the DPPC SLB was solubilised and progressively removed from the surface [43].

### *4.1.3. Langmuir-Blodgett and Langmuir-Schaefer deposition techniques*

of pure POPC, the mixtures of POPC/POPS or cardiolipin. The effect of membrane fluidity (acyl chain saturation), head group separation and the electrostatic shielding of the phospho‐ lipids strongly influenced the association of α-synuclein onto the membranes [34]. An analogue study was carried out on prion protein, which is widely recognised as the causative agent for Parkinson's disease when the protein is misfolded [33]. The interactions of N-terminal cleavage fragments (N1 and N2) of prion protein on negatively charged POPG-contained SLBs were probed using neutron reflectometry. The results implied that both N1 and N2 were inserted into phospholipid head groups; however, the interactions were stronger for N1 fragment because of the additional metal ion binding site located at a polybasic region. The insertion of N1 fragment induced an increase in lipid order in the absence of phase transition, as evidenced by lengthening of lipid acyl tails and decreasing in lipid area. It is possible that prion protein N-terminal fragment plays a functional role on membrane interactions [33]. Recently, Lu et al. [35] employed neutron reflectometry to define the molecular determinants of cellular protein layers adsorbed on mimetic apoptotic membranes. The interactions between annexin V and membranes contained different amount of phosphatidylserine (PS) phospholipids were examined, which is a crucial process to detect early/mid-stage cell membrane apoptosis *in vivo* and *in vitro*. The membrane properties that were affected by PS contents strongly affected annexin V binding, showing an increase in annexin V binding with an increase in PS content in the membrane and that there was a maximum binding at 20% of PS in the membrane. A decrease in the thickness of the adsorbed annexin V layer was observed with a 33% of PS content in membrane, suggesting the annexin V might alter its conformational change at high

Lipid-detergent method is an alternative method for depositing SLBs on solid support. Tiberg et al. [37, 38] have successfully demonstrated the formation of SLB with the aid of detergent incorporated into the lipid forming micelles. The presence of detergents allows saturated fatty acids such as DPPC that has high hydrophobicity, to form stable micelles in aqueous solution [39]. In this method, the detergent is first incorporated into the lipid solution in excess amount, this helps to emulsify the lipid molecules and raise to the critical micelle concentration. Once lipid molecules concentration reached the critical micelle concentration, they aggregated into stable lipid micelles, which were introduced and adsorbed onto a hydrophilic silica surface [37–40]. The resulted SLB was identical to those formed using the vesicle fusion method once the detergent was completely removed [29, 38, 40, 41]. The cholesterol incorporated DPPC membranes were also used to study the effect of added components on the resulting SLBs. The neutron reflectometry results showed that the cholesterol was located below the lipid head

group region and subsequently led to an increase in membrane thickness [38].

In addition to the membrane structural information, neutron reflectometry was also used to study protein interactions with membranes [29, 42]. Vacklin et al. [42] investigated the physical interactions between phospholipase A2 (PLA2) upon regulating its enzymatic hydrolysis on supported lipid bilayer. They were able to monitor the membrane bilayer changing its morphology when exposed to PLA2 and the reaction was irreversible. By using phospholipids

PS content [35].

134 Neutron Scattering

*4.1.2. Lipid-detergent method*

Langmuir-Blodgett and *Langmuir-Schaefer* deposition techniques are the most versatile methods for the preparation of thin and highly organised molecular films [44]. The amphiphilic lipid molecules have the self-assembled properties, which allow the lipid molecules to arrange in a monolayer film at an air/liquid interface on a solid support. The vertical lipid deposition usually forms a homogeneous layer with a controllable thickness of the deposited film, also called Langmuir-Blodgett film [45]. The solid support is placed vertically into the lipid solution. The resulted monolayer (first layer) has the lipid head groups interacting with the solid support and the acyl tail regions pointing towards the gas phase on a hydrophilic solid support, whereas on a hydrophobic solid support, the acyl tail regions interact with the solid surface, leaving the lipid head groups point outwards. Another similar method, *Langmuir-Schaefer* deposition in which the solid support is oriented horizontally is commonly used for lipid deposition [45]. Combining both vertical and horizontal lipid deposition techniques, the formation of SLB can be easily achieved using the Langmuir-Blodgett method for depositing the first lipid layer, and the *Langmuir-Schaefer* method for the deposition of the second lipid layer [45].

Recently, Clifton et al. [46] demonstrated how the Langmuir-Blodgett and Langmuir-Schaefer methods can be used to create a asymmetric SLB as a mimic of the complex Gram-negative bacterial outer membrane. The inner membrane leaflet of the outer membrane is mainly composed of phosphatidylcholine, which was deposited on hydrophilic surface using the Langmuir-Blodgett method. The outer membrane leaflet of outer membrane was deposited via Langmuir-Schaefer method, which contained either Lipid A or *Escherichia coli* rough lipopolysaccharides [46]. In the combination of both Langmuir-Blodgett and Langmuir-Schaefer techniques, the asymmetric SLBs were detected with greater than 90% surface coverage on solid support. The asymmetric SLBs also maintained high stability over time; this enabled us to clearly resolve the structure of the asymmetrical bilayer leaflets [46]. The development of functionalised biomimetic membrane using Langmuir-Blodgett and Lang‐ muir-Schaefer methods is of great interest in nanobiotechnology for years [45]. However, the applications of neutron reflectometry to study biological phenomenon on Langmuir-Blodgett and Langmuir*-*Schaefer method–formed SLBs are not available.

### **4.2. Tethered lipid bilayer membrane**

Tethered lipid bilayer membrane (t-BLM) is the lipid bilayer deposited on molecular anchors that act as spacers between the membrane and the solid support [47]. The use of self-assembled molecules as a surface coating technique has been developed since the early 1990s [47]. Molecules or ligands that form spontaneously via chemisorption from a solution or gas onto a solid support can result in a self-assembled monolayer [47]. Short oligomers or alkanethiols are often used as the applications of self-assembled monolayer that forms covalent bonds to the solid support such as gold and mercury, and the terminal functional groups create new surface properties on the solid support [15, 47, 48]. The deposition of lipid vesicles onto selfassembled monolayer allows the formation of tethered lipid bilayer membrane in which the lower bilayer leaflet interacts with the terminal functional group on self-assembled monolayer and the upper leaflet of phospholipids forms a functional lipid layer (Figure 3b) [15]. Tailoring the surface properties of the solid support improves the long-term stability for t-BLM, and hence widening the potential applications.

A neutron reflectometry study from Junghans et al. [49] used the selection of spacer group in affecting the formation of tethered lipid bilayer membrane. The types and the length of the spacers influenced the molecular geometry of the lipid molecules and hence affecting the structural and electrical properties on the membranes [49]. The use of long alkyl chain spacers increased water incorporation within the spacer units, which gave rise to a less dense packing order, an increase in both bilayer defect rate and roughness of the bilayer interfaces [49]. Alternatively, shorter spacer units had the highest surface coverage and also little water incorporation, resulting in an increase in membrane resistances [49]. With the controlled length of spacer, an optimum membrane architecture can be created, which allows for creating the physiological environment in which the protein-membrane interactions can be studied and further leading to possible biological applications.

Recently, the applications of tethered lipid bilayer membrane and neutron reflectometry were employed to investigate the orientations and conformations of membrane-bound protein, glucocerebrosidase (GCase), in an aqueous environment [50]. GCase was partially inserted into the tethered lipid bilayer membranes with its active site exposed to the membrane-water interface [50]. The use of selective deuteration techniques and the reflection of neutrons simultaneously extended the applications for protein complex characterisation. The results obtained from neutron reflectometry showed a large conformational change in GCase when interacted with α-synuclein (α-syn) [50]. Previous research has suggested that the formation of α-syn/GCase complex inhibited the enzymatic reaction of GCase, which might lead to the development of Parkinson disease and related disorders [51].

### **4.3. Polymer cushioned membranes**

Polymer cushioned membranes are lipid bilayers deposited on a soft polymeric layer that coated on a solid support (Figure 3c) [16]. The use of polymeric layer mimics the cytoskeleton or extracellular matrix in plasma membrane, which is important in reconstituting the mor‐ phology of membrane domains as well as maintaining the nature and mobility of the trans‐ membrane proteins [16, 29]. The polymeric layer must have the following properties: the hydrophilic polymer is able to form a thin, uniform layer for the deposition of lipid bilayer. And secondly, the polymeric layer has a well-defined elastic modulus that is highly reprodu‐ cible and chemically inert to the membranes or to the solid support [16]. The thickness, density and water content of the polymer cushioned layer as well as the membrane bilayer can be determined by neutron reflectometry. The deformability of polymer cushioned membranes provides a more natural cell-like environment for membrane-associated process [52]. The experimental setup of neutron reflectometry allows the lipid bilayer to maintain at a physio‐ logical condition where the membrane exists in the fluid state. This provides a great potential to study and the membrane-mediated interactions and monitor the experimental conditions *in situ* [53].

The use of beta-amyloid peptides as the model for Alzheimer's disease has been widely studied and recognised [54]. Dante et al. [53] used neutron reflectometry to study the role of betaamyloid peptides on polymer cushioned membranes. The data from neutron reflectometry evidenced a change in the lateral diffusion of the lipid bilayer when exposed to beta-amyloid peptides. It, therefore, suggested that the beta-amyloid peptides penetrated into the lipid bilayer and softened the lipid membrane bilayers [53]. These results strongly indicated that the change in lipid membranes' structural and mechanical properties could influence and alter membrane-based signal transduction. Monitoring the interactions of beta-amyloid peptides on membrane with neutron reflectometry provided insights into the onset of Alzheimer's disease [53].

### **4.4. Floating supported bilayers**

**4.2. Tethered lipid bilayer membrane**

136 Neutron Scattering

hence widening the potential applications.

further leading to possible biological applications.

development of Parkinson disease and related disorders [51].

**4.3. Polymer cushioned membranes**

Tethered lipid bilayer membrane (t-BLM) is the lipid bilayer deposited on molecular anchors that act as spacers between the membrane and the solid support [47]. The use of self-assembled molecules as a surface coating technique has been developed since the early 1990s [47]. Molecules or ligands that form spontaneously via chemisorption from a solution or gas onto a solid support can result in a self-assembled monolayer [47]. Short oligomers or alkanethiols are often used as the applications of self-assembled monolayer that forms covalent bonds to the solid support such as gold and mercury, and the terminal functional groups create new surface properties on the solid support [15, 47, 48]. The deposition of lipid vesicles onto selfassembled monolayer allows the formation of tethered lipid bilayer membrane in which the lower bilayer leaflet interacts with the terminal functional group on self-assembled monolayer and the upper leaflet of phospholipids forms a functional lipid layer (Figure 3b) [15]. Tailoring the surface properties of the solid support improves the long-term stability for t-BLM, and

A neutron reflectometry study from Junghans et al. [49] used the selection of spacer group in affecting the formation of tethered lipid bilayer membrane. The types and the length of the spacers influenced the molecular geometry of the lipid molecules and hence affecting the structural and electrical properties on the membranes [49]. The use of long alkyl chain spacers increased water incorporation within the spacer units, which gave rise to a less dense packing order, an increase in both bilayer defect rate and roughness of the bilayer interfaces [49]. Alternatively, shorter spacer units had the highest surface coverage and also little water incorporation, resulting in an increase in membrane resistances [49]. With the controlled length of spacer, an optimum membrane architecture can be created, which allows for creating the physiological environment in which the protein-membrane interactions can be studied and

Recently, the applications of tethered lipid bilayer membrane and neutron reflectometry were employed to investigate the orientations and conformations of membrane-bound protein, glucocerebrosidase (GCase), in an aqueous environment [50]. GCase was partially inserted into the tethered lipid bilayer membranes with its active site exposed to the membrane-water interface [50]. The use of selective deuteration techniques and the reflection of neutrons simultaneously extended the applications for protein complex characterisation. The results obtained from neutron reflectometry showed a large conformational change in GCase when interacted with α-synuclein (α-syn) [50]. Previous research has suggested that the formation of α-syn/GCase complex inhibited the enzymatic reaction of GCase, which might lead to the

Polymer cushioned membranes are lipid bilayers deposited on a soft polymeric layer that coated on a solid support (Figure 3c) [16]. The use of polymeric layer mimics the cytoskeleton or extracellular matrix in plasma membrane, which is important in reconstituting the mor‐ phology of membrane domains as well as maintaining the nature and mobility of the trans‐ membrane proteins [16, 29]. The polymeric layer must have the following properties: the Floating supported bilayers were first described by Fragneto et al. [55] as a new artificial membrane system formed on a solid support. In this SLB model, the lipid bilayer was formed in two steps: a Langmuir-Blodgett technique and a Langmuir-Schaefer technique. Firstly, a Langmuir-Blodgett technique was used to deposit a monolayer of lipid onto the solid support that was placed vertically (lipid monolayer with dark blue headgroup on Figure 3d). A lipid bilayer (lipids with light blue headgroup) was then horizontally introduced onto the lipid monolayer using a Langmuir-Schaefer technique (i.e., horizontal deposition). Therefore, the second bilayer interacted with the lipid monolayer, hence named floating lipid bilayer (lipid bilayer with light blue headgroup on Figure 3d) [55].

Talbot and co-workers [41] used neutron reflectivity technique to compare the formation of SLBs using various methods as described previously, this was further extended to investigate the interactions between a cationic gene delivery vectors (lipoplexes) with different SLB model membranes such as direct membrane deposition and polymer cushioned membranes. The data showed that the cationic lipoplexes could easily destroy the floating supported bilayers on silicon dioxide surface as well as those SLB formed on polymer-cushioned silicon surface. The surface of the solid support might exert some influence to the destruction of floating supported bilayers. However, a chemically grafted membrane layer, octadecyltrichlorosilane (OTS) hybrid floating supported bilayer, is a perfect model bilayer to investigate the interactions between lipoplex and membranes. This membrane model had proven high stability to the exposure of solvents and lipoplexes over a considerable duration of time, suggesting the OTS- floating supported bilayers were not destroyed by the lipoplexes. Their work has established a new floating supported bilayers model that has minimum interference from the solid support, and hence it is suitable for studying the lipoplex-membrane interactions and corre‐ lates the transfection efficiency of this gene-delivery vector [41].

The usefulness of floating supported bilayers as a physiological mimic to the bacterial outer membrane helps to encounter technical challenge to recreate and study the interactions of antimicrobial proteins *in vitro*. Clifton et al. [56] described the formation of asymmetric floating supported bilayers fabricated by sequential Langmuir-Blodgett deposition of 1-oleoyl-2-(16 thiopalmitoyl)-sn-glycero-3-phosphocholine (thiolPC) and Langmuir-Schaefer deposition of floating supported bilayers consisted of lipopolysaccharides to mimic the bacterial outer membrane. The thiolPC layer and the floating supported bilayers are separated by a small water gap; hence the DPPC and lipopolysaccharides bilayers are floating on the solid support [56]. This floating-supported bilayer system has been fully studied and characterised using neutron reflectometry with high stability and reproducibility. The use of floating supported bilayers makes it possible to maintain the dynamicity, fluidity and translation freedom of the membranes; therefore, the membrane properties are not altered and behave similarly in their physiological environment [56–58]. Further studies show that lactoferrin, a cationic antimi‐ crobial protein, directly interacts with lipopolysaccharides on the outer membrane via electronic interactions. The results from neutron reflectometry revealed a dramatic reduction in the floating supported bilayer coverage in the presence of lactoferrin, providing a structural picture of lactoferrin disrupting the outer membrane [56]. Another cationic enzyme, lysozyme was found to bind to the outer membrane electrostatically without changing the floating supported bilayer coverage, and a small increase in membrane bilayer roughness was also detected. These two studies both reported a change in floating supported bilayers properties upon interactions with the antimicrobial proteins; however, the neutron reflectometry data implied that the lysozyme disrupted the outer membrane in a different mechanism than lactoferrin [56].

### **5. Membrane proteins on gold-modified surface**

Membrane proteins are well-known to be difficult to handle and study outside their natural lipid environment because they are flexible, unstable and relatively hydrophobic [1]. To overcome the problems, membrane proteins are extracted from membranes in the existence of detergents, which helps to stabilise membrane proteins during extraction, solubilisation, purification and crystallisation [1, 59]. To examine the structure-function relationships of membrane proteins, reconstitution of the artificial membrane environment is also essential in maintaining native protein configurations for biological functions [60]. Unfortunately, the detergents, which stabilise the membrane proteins, can easily disrupt the lipid bilayers (as described in Section 4) by penetrating the membranes and then form lipid micelles. Researchers are motivated to develop new approaches to encounter the technical difficulties in reconsti‐ tuting the native environment for studying membrane proteins *in vitro* [61].

As with any structural or functional investigations with membrane proteins, the use of direct protein immobilisation on a solid support has been used intensively in the protein array and biosensors industry [62, 63]. Often the challenge for this method is that the immobilised proteins adapt into random orientations, causing a change in protein conformations and consequently reducing the bioactivity. Therefore, conformationally oriented immobilisation strategy should be considered to enhance the stability of the immobilised proteins. Sitedirected mutagenesis, a protein-engineering method, is particularly useful to improve the quality of protein immobilisation on a solid support. The selected amino acid sequences of protein are precisely mutated and the modified amino acids are used for surface immobilisa‐ tion [62, 63]. Two different approaches, cysteine and histidine residues mutations, have been developed as illustrated in **Figure 4** [63–66]. The designed mutations (either cysteine or histidine) allow the membrane proteins to immobilise with a particular orientation in which their native conformations and activities are retained upon adsorption [14, 27, 67, 68]. Suc‐ cessfully, control protein orientation on a modified surface is a key to investigate the interac‐ tions and stereochemistry of multi-protein system [67].

floating supported bilayers were not destroyed by the lipoplexes. Their work has established a new floating supported bilayers model that has minimum interference from the solid support, and hence it is suitable for studying the lipoplex-membrane interactions and corre‐

The usefulness of floating supported bilayers as a physiological mimic to the bacterial outer membrane helps to encounter technical challenge to recreate and study the interactions of antimicrobial proteins *in vitro*. Clifton et al. [56] described the formation of asymmetric floating supported bilayers fabricated by sequential Langmuir-Blodgett deposition of 1-oleoyl-2-(16 thiopalmitoyl)-sn-glycero-3-phosphocholine (thiolPC) and Langmuir-Schaefer deposition of floating supported bilayers consisted of lipopolysaccharides to mimic the bacterial outer membrane. The thiolPC layer and the floating supported bilayers are separated by a small water gap; hence the DPPC and lipopolysaccharides bilayers are floating on the solid support [56]. This floating-supported bilayer system has been fully studied and characterised using neutron reflectometry with high stability and reproducibility. The use of floating supported bilayers makes it possible to maintain the dynamicity, fluidity and translation freedom of the membranes; therefore, the membrane properties are not altered and behave similarly in their physiological environment [56–58]. Further studies show that lactoferrin, a cationic antimi‐ crobial protein, directly interacts with lipopolysaccharides on the outer membrane via electronic interactions. The results from neutron reflectometry revealed a dramatic reduction in the floating supported bilayer coverage in the presence of lactoferrin, providing a structural picture of lactoferrin disrupting the outer membrane [56]. Another cationic enzyme, lysozyme was found to bind to the outer membrane electrostatically without changing the floating supported bilayer coverage, and a small increase in membrane bilayer roughness was also detected. These two studies both reported a change in floating supported bilayers properties upon interactions with the antimicrobial proteins; however, the neutron reflectometry data implied that the lysozyme disrupted the outer membrane in a different mechanism than

Membrane proteins are well-known to be difficult to handle and study outside their natural lipid environment because they are flexible, unstable and relatively hydrophobic [1]. To overcome the problems, membrane proteins are extracted from membranes in the existence of detergents, which helps to stabilise membrane proteins during extraction, solubilisation, purification and crystallisation [1, 59]. To examine the structure-function relationships of membrane proteins, reconstitution of the artificial membrane environment is also essential in maintaining native protein configurations for biological functions [60]. Unfortunately, the detergents, which stabilise the membrane proteins, can easily disrupt the lipid bilayers (as described in Section 4) by penetrating the membranes and then form lipid micelles. Researchers are motivated to develop new approaches to encounter the technical difficulties in reconsti‐

tuting the native environment for studying membrane proteins *in vitro* [61].

lates the transfection efficiency of this gene-delivery vector [41].

**5. Membrane proteins on gold-modified surface**

lactoferrin [56].

138 Neutron Scattering

**Figure 4.** Protein immobilisation on gold (Au) or gold-modified surfaces: (a) direct immobilisation allows proteins ori‐ ented randomly, however, proteins orientation can be controlled by (b) thiol chemistry on cysteine residues (red) or (c) histidine (blue) interactions with nickel-nitrilotriacetic acid (Ni-NTA)-modified gold surfaces.

### **5.1. Substituted cysteine in protein forms a gold-thiolate covalent bond**

The cysteine-containing proteins can be directly immobilised on gold surfaces using thiol chemistry [69], i.e., the sulphur atom on the cysteine residue forms a strong gold-thiolate covalent bond with the gold surfaces [12]. The spontaneous deposition promotes the formation of a dense self-assembled monolayer on gold. The application of cysteine-thiol immobilisation is commonly associated with biosensors and arrays applications. Brun et al. [12] used bacterial outer membrane proteins as a model for protein assay, which has potential applications as scaffolds for tissue engineering, proteomics and diagnostics. The outer membrane protein A (OmpA) is a monomeric membrane protein that involves in the survival mechanism in *Escherichia coli* (*E. coli*) when exposed to environmental stresses. The beta-barrel transmem‐ brane structure of OmpA spans though the bacterial outer membranes and plays a vital role in membrane stability [70]. In Brun and co-workers' studies, OmpA was engineered by inserting a cysteine residue into the periplasmic turn, such that the OmpA was immobilised on gold surfaces in a specific orientation via the covalent gold-thiolate bond. The immobilised OmpA was circularly permutated in order to expose the c and n termini for the interactions with *Staphylococcus aureus* protein A (SpA), a pathogen bound to bacterium cell wall. The SpA bound to OmpA using a mutated B domain (named as Z domain) created an OmpAZ protein scaffold on the gold surface [70]. Neutron reflectometry was used to detect the addition of Immunoglobulin G (IgG) on the OmpAZ protein scaffold, and the thickness of each protein layers was determined in high resolution. This provided information on the orientation of IgG on the OmpAZ protein scaffold [70]. Furthermore, Brun and co-workers enhanced the detection of IgG by introducing a flexible linker as a functional motif to link two Z domains together on OmpA (ZZOmpA). The capacity of IgG binding to ZZOmpA was greatly enhanced and hence potentially increased the signal and sensitivity for protein array applications [71, 72]. The antibody protein array (IgG-ZZOmpA) has been further used to detect antigen influenza nucleoprotein (NP), that is highly conserved and the basis of diagnostic tests for influenza. The NP binding capacity as well as the structural orientation of the bound antibodyantigen layer, ZZOmpA-NP layer, were fully characterised [65]. Brun et al. [65] have demon‐ strated the potential of using neutron reflectometry as an *in situ* physical characterisation of the protein assay assembly process. The cysteine-orientated immobilisation technique enables the proteins to be used to maintain in their native structure in order to be fully functional as a protein scaffold for further applications such as antibody and antigen detections.

### **5.2. Histidine-tagged membrane protein binds to nickel-nitrilotriacetic acid**

Another protein immobilisation strategy is based on the covalent interactions between histidine (His) residues and nickel-nitrilotriacetic acid (Ni-NTA) modified surfaces. The His residues are genetically engineered on either the N or the C terminus of recombinant proteins [63]. The nickel-nitrilotriacetic acid (Ni-NTA) functionalised gold surface has a high affinity to capture His-tagged proteins, and hence all bound proteins are conferred in a particular orientation [67]. The His-Ni-NTA affinity technique enables the proteins immobilise on a solid surface without altering their function, which has been a challenge in the development of protein assay for decades [63, 67]. The use of recombinant affinity tags has addressed the issues of protein orientation and surface density upon immobilisation on solid surface [67, 73].

Shen et al*.* [63] have reconstituted an artificial environment for the translocation and assembly module (Tam) nanomachine that catalyses the insertion and assembly of nascent membrane proteins into the outer membranes of bacteria. The TAM nanomachine is composed of two proteins: a beta-barrel contained TamA locates at the outer bacterial membrane and a betahelical TamB spans in the inner bacterial membrane. The two protein subunits, TamA and TamB, are linked together by a three polypeptide-transport-associated (POTRA) domains in TamA that is located in the periplasm region. The precise movement and activity of the TAM nanomachines were studied using neutron reflectometry to provide an accurate measurement on the molecular movement of the Tam protein complexes [63]. Firstly, a hexa-histidine tag was specifically engineered into an extracellular loop of TamA to immobilise onto the Ni-NTA functionalised gold surface, followed by the addition of lipids for membrane reconstitution. The thickness of the membrane embedded beta-barrel, the three POTRA domains as well as the hexahistidine extramembrane layers were separately determined in the reconstituted membrane environment. The architecture of the TAM complex (TamA and TamB) was also studied using neutron reflectometry; no structural movement was obtained upon interactions of TamB to the POTRA domains of TamA [63]. This study focuses on the interactions between a substrate protein, Ag43, and the TAM complex on a supported membrane. The presence of Ag43 triggered the POTRA domain of TamA to alter its conformational changes, projected away from the membrane. The movement of TamA created spatial constraint on the outer membrane and influenced TamB deposition within the Tam complex. Monitoring the TAM complex assembly pathways using neutron reflectometry allows further understanding on the operations of diverse cellular processes for virulence of bacterial pathogens [63].

### **6. Concluding remarks**

on gold surfaces in a specific orientation via the covalent gold-thiolate bond. The immobilised OmpA was circularly permutated in order to expose the c and n termini for the interactions with *Staphylococcus aureus* protein A (SpA), a pathogen bound to bacterium cell wall. The SpA bound to OmpA using a mutated B domain (named as Z domain) created an OmpAZ protein scaffold on the gold surface [70]. Neutron reflectometry was used to detect the addition of Immunoglobulin G (IgG) on the OmpAZ protein scaffold, and the thickness of each protein layers was determined in high resolution. This provided information on the orientation of IgG on the OmpAZ protein scaffold [70]. Furthermore, Brun and co-workers enhanced the detection of IgG by introducing a flexible linker as a functional motif to link two Z domains together on OmpA (ZZOmpA). The capacity of IgG binding to ZZOmpA was greatly enhanced and hence potentially increased the signal and sensitivity for protein array applications [71, 72]. The antibody protein array (IgG-ZZOmpA) has been further used to detect antigen influenza nucleoprotein (NP), that is highly conserved and the basis of diagnostic tests for influenza. The NP binding capacity as well as the structural orientation of the bound antibodyantigen layer, ZZOmpA-NP layer, were fully characterised [65]. Brun et al. [65] have demon‐ strated the potential of using neutron reflectometry as an *in situ* physical characterisation of the protein assay assembly process. The cysteine-orientated immobilisation technique enables the proteins to be used to maintain in their native structure in order to be fully functional as a

140 Neutron Scattering

protein scaffold for further applications such as antibody and antigen detections.

Another protein immobilisation strategy is based on the covalent interactions between histidine (His) residues and nickel-nitrilotriacetic acid (Ni-NTA) modified surfaces. The His residues are genetically engineered on either the N or the C terminus of recombinant proteins [63]. The nickel-nitrilotriacetic acid (Ni-NTA) functionalised gold surface has a high affinity to capture His-tagged proteins, and hence all bound proteins are conferred in a particular orientation [67]. The His-Ni-NTA affinity technique enables the proteins immobilise on a solid surface without altering their function, which has been a challenge in the development of protein assay for decades [63, 67]. The use of recombinant affinity tags has addressed the issues of protein orientation and surface density upon immobilisation on solid surface [67, 73].

Shen et al*.* [63] have reconstituted an artificial environment for the translocation and assembly module (Tam) nanomachine that catalyses the insertion and assembly of nascent membrane proteins into the outer membranes of bacteria. The TAM nanomachine is composed of two proteins: a beta-barrel contained TamA locates at the outer bacterial membrane and a betahelical TamB spans in the inner bacterial membrane. The two protein subunits, TamA and TamB, are linked together by a three polypeptide-transport-associated (POTRA) domains in TamA that is located in the periplasm region. The precise movement and activity of the TAM nanomachines were studied using neutron reflectometry to provide an accurate measurement on the molecular movement of the Tam protein complexes [63]. Firstly, a hexa-histidine tag was specifically engineered into an extracellular loop of TamA to immobilise onto the Ni-NTA functionalised gold surface, followed by the addition of lipids for membrane reconstitution. The thickness of the membrane embedded beta-barrel, the three POTRA domains as well as

**5.2. Histidine-tagged membrane protein binds to nickel-nitrilotriacetic acid**

Understanding the protein-membrane interactions is critical to expanding the applications in drug delivery, biological engineering and, especially, biosensors and microarrays [4]. In this chapter, examples of protein adsorption on various artificial membrane models are reviewed and discussed. Along with the technical advances of neutron reflectometry, information on molecular binding and environment and orientation of the lipids and biomolecules is inves‐ tigated in great detail. The use of various artificial membrane models offers a methodological design suited to study protein/peptide-membrane interactions in different biological systems. The phospholipid monolayer models are particularly useful for biological applications exploring the influence of fluidity or surface properties [74]. The phospholipid monolayer consists of one leaflet of lipid, which represents only a half bilayer. The structural simplicity and versatility allow for monitoring accurately the lipid packing density [74]. Traditionally, the phospholipid monolayer is deposited at an air/water interface using the Langmuir-Blodgett method; the surface coverage of lipid monolayer at air/water interface can be easily controlled, enabling examination of the influence of phospholipid membrane properties on protein- and peptide-membrane interactions [18]. Alternatively, lipid can be deposited and functionalised using a polymer or surfactant at a solid/liquid interface [17]. The increasing availability of different artificial membrane models extends and improves the ability to mimic the biological equivalent.

Protein adsorption studies on various fluid phospholipid bilayers have been developed: (a) single supported lipid bilayers, (b) tethered lipid bilayer membranes, (c) polymer-cushioned bilayer membranes and (d) floating supported bilayers [29]. The ease of preparation and high stability of the support lipid bilayer make it the most popular artificial membrane model [29]. Tethered lipid bilayer membranes are complex interfacial structures where the phospholipid bilayers are deposited on molecular anchors that are chemically linked between the membrane and the solid support [47]. These tethered lipid bilayer membranes provide a thicker inter‐ mediate water layer that mimics the natural cell–like environment for membrane-associated process [74]. Fluid lipid membranes can be deposited on a soft, polymeric layer, named polymer cushioned membranes [16]. The incorporation of molecular anchors and polymers between membranes and solid surface improves the long-term stability of the phospholipids and provides a better biomimetic condition for the applications of membrane proteins [53]. Recently, floating supported bilayers are used as an advance membrane model to mimic the phase behaviour of lipids and the fluidity of bilayers in nature. The floating supported bilayers are created using a series of both Langmuir-Blodgett and Langmuir-Schaefer methods onto a solid support, and this is the most suitable system to replicate the asymmetry of bacterial outer membranes [55]. However, artificial membranes may not always be the most suitable system to investigate and examine the structural and functional aspects of membrane proteins because (a) lipid bilayers can be destroyed by the membrane protein solution that contains surfactant and (b) it is difficult to control the protein orientation on a solid support [60]. Thus, membrane proteins can be modified or mutated at desired locations to control the protein orientation when immobilised on gold-modified surfaces. Cysteine or histidine residues are the common approaches for the binding of protein onto thiol-modified or nickel-nitrilotriacetic acidmodified gold surfaces [63–66]. These conformationally oriented immobilisation strategies allow the membrane proteins to retain their native conformations and activities upon adsorp‐ tion and also it creates a stable lipid layer surrounding the membrane protein. This review illustrates and highlights examples of the role of neutron reflectometry for studying proteins and peptides in biomimetic membranes. Many important advantages and unique features of neutron reflectometry offer numerous opportunities for innovative applications and technical development for *in vitro* study of biomolecules. Only a small number of neutron reflectometry applications in protein adsorption are demonstrated, the potential for studying proteinmembrane interactions in multi-components should not be underestimated in the future [15].

### **Author details**

Joyee Chun In Yeung1 , Tsung-Wu Lin2 and Hsin-Hui Shen1\*

\*Address all correspondence to: hsin-hui.shen@monash.edu

1 Infection and Immunity Program, Biomedicine Discovery Institute and Department of Microbiology, Monash University, Melbourne, Australia

2 Department of Chemistry, Tunghai University, Taichung City, Taiwan

### **References**


[3] Yu Q, Chen H. Interaction of switchable biomaterials surfaces with proteins. In: Switchable and Responsive Surfaces and Materials for Biomedical Applications. Woodhead Publishing, 2015.

and provides a better biomimetic condition for the applications of membrane proteins [53]. Recently, floating supported bilayers are used as an advance membrane model to mimic the phase behaviour of lipids and the fluidity of bilayers in nature. The floating supported bilayers are created using a series of both Langmuir-Blodgett and Langmuir-Schaefer methods onto a solid support, and this is the most suitable system to replicate the asymmetry of bacterial outer membranes [55]. However, artificial membranes may not always be the most suitable system to investigate and examine the structural and functional aspects of membrane proteins because (a) lipid bilayers can be destroyed by the membrane protein solution that contains surfactant and (b) it is difficult to control the protein orientation on a solid support [60]. Thus, membrane proteins can be modified or mutated at desired locations to control the protein orientation when immobilised on gold-modified surfaces. Cysteine or histidine residues are the common approaches for the binding of protein onto thiol-modified or nickel-nitrilotriacetic acidmodified gold surfaces [63–66]. These conformationally oriented immobilisation strategies allow the membrane proteins to retain their native conformations and activities upon adsorp‐ tion and also it creates a stable lipid layer surrounding the membrane protein. This review illustrates and highlights examples of the role of neutron reflectometry for studying proteins and peptides in biomimetic membranes. Many important advantages and unique features of neutron reflectometry offer numerous opportunities for innovative applications and technical development for *in vitro* study of biomolecules. Only a small number of neutron reflectometry applications in protein adsorption are demonstrated, the potential for studying proteinmembrane interactions in multi-components should not be underestimated in the future [15].

and Hsin-Hui Shen1\*

[1] Carpenter EP, Beis K, Cameron AD, Iwata S. Overcoming the challenges of mem‐ brane protein crystallography. Current Opinion in Structural Biology. 2008;18(5):

[2] Tanaka M. Supported membranes – Structure and interactions. In: Comprehensive

1 Infection and Immunity Program, Biomedicine Discovery Institute and Department of

**Author details**

142 Neutron Scattering

**References**

581-586.

Joyee Chun In Yeung1

, Tsung-Wu Lin2

\*Address all correspondence to: hsin-hui.shen@monash.edu

Microbiology, Monash University, Melbourne, Australia

2 Department of Chemistry, Tunghai University, Taichung City, Taiwan

Biophysics: Elsevier. 2012. pp 261-272. ISBN:9780080957180


[28] Schmidt A, Spinke J, Bayerl T, Sackmann E, Knoll W. Streptavidin binding to biotiny‐ lated lipid layers on solid supports. A neutron reflection and surface plasmon optical study. Biophysical Journal. 1992;63:1185-1192. DOI:10.1016/S0006-3495(92)81715-0

[17] Li JB, He Q, Yan Xh. Biomimetic Membranes. In: *Molecular Assembly of Biomimetic*

[18] Pirrone GF, Vernon BC, Michael S. Kent, Engen JR. Hydrogen exchange mass spec‐ trometry of proteins at Langmuir monolayers. Analytical Chemistry.

[19] Pichot R, Watson RL, Norton IT. Phospholipids at the interface: Current trends and challenges. International Journal of Molecular Sciences. 2013;14:11767-11794. DOI:

[20] Maierhofer AP, Bucknall DG, Bayerl TM. Modulation of cytochrome c coupling to anionic lipid monolayers by a change of the phase state: A combined neutron and in‐ frared reflection study. Biophysical Journal. 2000;79:1428-1437. DOI:10.1016/

[21] Miano F, Zhao X, Lu JR, Penfold J. Coadsorption of human milk lactoferrin into the dipalmitoylglycerolphosphatidylcholine phospholipid monolayer spread at the air/ water interface. Biophysical Journal. 2007;92:1254-1262. DOI:10.1529/biophysj.

[22] Pirrone GF, Emert-Sedlak LA, Wales TE, Smithgall TE, Kent MS, Engen JR. Mem‐ brane-associated conformation of HIV-1 Nef investigated with hydrogen exchange mass spectrometry at a Langmuir monolayer. Analytical Chemistry.

[23] Kuhl TL, Majewski J, Wong JY, Steinberg S, Leckband DE, Israelachvili JN, et al. A neutron reflectivity study of polymer-modified phospholipid monolayers at the sol‐ id-solution interface: Polyethylene glycol-lipids on silane-modified substrates. Bio‐

[24] Hollinshead CM, Hanna M, Barlow DJ, Biasi VD, Bucknall DG, Camilleri P, et al. Neutron reflection from a dimyristoylphosphatidylcholine monolayer adsorbed on a hydrophobised silicon support. Biochimica et Biophysica Acta. 2001;1511:49-59. DOI:

[25] Penfold J, Richardson RM, Zarbakhsh A, Webster JRP, Bucknall DG, Rennie AR, et al. Recent advances in the study of chemical surfaces and interfaces by specular neu‐ tron reflection. Journal of the Chemical Society, Faraday Transactions. 1997;93(22):

[26] Lu JR, Murphy EF, Su TJ. Reduced protein adsorption on the surface of a chemically grafted phospholipid monolayer. Langmuir. 2001;17:3382-3389. DOI:10.1021/

[27] Lu JR, Zhao X, Yaseen M. Protein adsorption studied by neutron reflection. Current Opinion in Colloid & Interface Science. 2007;12:9-16. DOI:10.1016/j.cocis.2007.02.001

physical Journal. 1998;75:2352-2362. DOI:10.1016/S0006-3495(98)77679-9

*Systems*. Wiley, 2011; pp. 7-39. ISBN:978-3-527-32542-9

2015;87:7022-7029. DOI:10.1021/acs.analchem.5b01724

2015;87:7030-7035. DOI:10.1021/acs.analchem.5b01725

10.3390/ijms140611767

S0006-3495(00)76394-6

10.1016/S0005-2736(00)00380-1

3899-3917. DOI:10.1039/A702836I

105.078592

144 Neutron Scattering

la0017429


[54] Murphy MP, III HL. Alzheimer's disease and the β-amyloid peptide. Journal of Alz‐ heimer's Disease. 2010;19(1). DOI:10.3233/JAD-2010-1221.

[42] Vacklin HP, Tiberg F, Fragneto G, Thomas RK. Phospholipase A2 hydrolysis of sup‐ ported phospholipid bilayers: A neutron reflectivity and ellipsometry study. Bio‐

[43] Shen H-H, Thomas RK, Penfold J, Fragneto G. Destruction and solubilization of sup‐ ported phospholipid bilayers on silica by the biosurfactant surfactin. Langmuir.

[44] Corkery RW. Langmuir−Blodgett (L−B) multilayer films. Langmuir. 1997;13(14):

[45] Girard-Egrot AP, Blum LJ. Langmuir-Blodgett technique for synthesis of biomimetic lipid membranes. In: Nanobiotechnology of Biomimetic Membranes. Fundamental

[46] Clifton LA, Skoda MWA, Daulton EL, Hughes AV, Brun APL, Lakey JH, et al. Asym‐ metric phospholipid: Lipopolysaccharide bilayers: A Gram-negative bacterial outer membrane mimic. Journal of Royal Society Interface. 2013;10(89). DOI:10.1098/rsif.

[47] Love JC, Estroff LA, Kriebel JK, Nuzzo RG, Whitesides GM. Self-assembled monolay‐ ers of thiolates on metals as a form of nanotechnology. Chemical Reviews.

[48] Knobler CM, Schwartz DK. Langmuir and self-assembled monolayers. Current Opin‐ ion in Colloid & Interface Science. 1999;4(1):46-51. DOI:10.1016/

[49] Junghans A, Koper I. Structural analysis of tethered bilayer lipid membranes. Lang‐

[50] Yap TL, Jiang Z, Heinrich F, Gruschus JM, Pfefferkorn CM, Barros M, et al. Structural features of membrane-bound glucocerebrosidase and α-synuclein probed by neutron reflectometry and fluorescence spectroscopy. Journal of Biological Chemistry.

[51] Douglas MR, Lewthwaite AJ, Nicholl DJ. Genetics of Parkinson's disease and parkin‐ sonism. Expert Review of Neurotherapeutics. 2007;7(6):657-666. DOI:

[52] Hertrich S, Stetter F, Rühm A, Hugel T, Nickel B. Highly hydrated deformable poly‐ ethylene glycol-tethered lipid bilayers. Langmuir. 2014;30:9442-9447. DOI:10.1021/

[53] Dante S, Hau T, Steitz R, Canale C, Dencher NA. Nanoscale structural and mechani‐ cal effects of beta-amyloid (1–42) on polymer cushioned membranes: A combined study by neutron reflectometry and AFM force spectroscopy. Biochimica et Biophysi‐

ca Acta. 2011;1808:1646-1655. DOI:10.1016/j.bbamem.2011.07.024.

chemistry. 2005;44:2811-2821. DOI:10.1021/la047389e

2010;26(10):7334-7342. DOI:10.1021/la904212x

Biomedical Technologies. 2007, pp. 23-74.

2005;105:1103-1169. DOI:10.1021/cr0300789

muir. 2010;26(13):11035-11040. DOI:10.1021/la100342k

2015;290(2):744-754. DOI:10.1074/jbc.M114.610584

3591-3594. DOI:10.1021/la9701743

2013.0810

146 Neutron Scattering

S1359-0294(99)00002-3

10.1586/14737175.7.6.657

la4045804


[67] Ley C, Holtmann D, Mangold K-M, Schrader J. Immobilization of histidine-tagged proteins on electrodes. Colloids and Surfaces B: Biointerfaces. 2011;88:539-551. DOI:

[68] Unsworth LD, Tun Z, Sheardown H, Brash JL. In situ neutron reflectometry investi‐ gation of gold-chemisorbed PEO layers of varying chain density: Relationship of lay‐ er structure to protein resistance. Journal of Colloid and Interface Science.

[69] Tao F, Bernasek SL. Self-assembled monolayers. In: *Comprehensive Nanoscience and*

[70] Wang Y. The function of OmpA in Escherichia coli. Biochemical and Biophysical Re‐ search Communications. 2002;292(2):396-401. DOI:10.1006/bbrc.2002.6657

[71] Brun APL, Shah DSH, Athey D, Holt SA, Lakey JH. Self-assembly of protein mono‐ layers engineered for improved monoclonal immunoglobulin G binding. Internation‐

[72] Brun APL, Holt SA, Shah DSH, Majkrzak CF, Lakey JH. The structure orientation of antibody layers bound to engineered biosensor surfaces. Biomaterials.

[73] Oshige M, Yumoto K, Miyata H, Takahashi S, Nakada M, Ito K, et al. Immobilization of His-tagged proteins on various solid surfaces using NTA-modified chitosan. Open

[74] Czolkos I, Jesorka A, Orwar O. Molecular phospholipid films on solid supports. Soft

Journal of Polymer Chemistry. 2013;3:6-10. DOI:10.4236/ojpchem.2013.31002

al Journal of Molecular Science. 2011;12:5157-5167. DOI:10.3390/ijms12085157

10.1016/j.colsurfb.2011.07.044

148 Neutron Scattering

2006;296(2):520-526. DOI:10.1016/j.jcis.2005.09.056

*Technology*. Elsevier, 2011. pp. 127-152. ISBN:978-0-12-374390-9

2011;32:3303-3311. DOI:10.1016/j.biomaterials.2011.01.026

Matter. 2011;7:4562-4576. DOI:10.1039/C0SM01212B

### *Edited by Waldemar Alfredo Monteiro*

This book brings suitable data concerning theory and experiments of neutron interactions with different materials. Since the neutron discovery by Chadwick in 1932, researchers of the entire world begin to make studies about it. It is well known that neutron have no charge, and their electric dipole moment is either zero or too small to measure, but theories and experiments show that neutron has spin (presence of magnetic moment), and polarization neutron scattering is plausible. The reader can obtain remarks about inelastic scattering cross sections for neutron; polarized neutron reflectivity; scattering methods; neutron reflectometry tool to probe the chemical structures; neutron scattering for amino acid crystals; and small-angle neutron scattering nanoemulsion heat transfer fluids in this book.

Photo by sakkmesterke / AdobeStock

Neutron Scattering

Neutron Scattering

*Edited by Waldemar Alfredo Monteiro*