飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

(1)

國

立

交

通

大

學

應用化學系分子科學碩士班

碩

士論文

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

Femtosecond Laser-Induced Crystallization of Amino Acid

and Protein at the Air/Solution Interface

研究生：劉宗翰 (Tsung-Han Liu)

指導教授：增原宏博士 (Dr. Hiroshi Masuhara)

(2)

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化

Femtosecond Laser-Induced Crystallization of Amino Acid

and Protein at the Air/Solution Interface

研究生：劉宗翰

Student：Tsung-Han Liu

指導教授：增原宏博士

Advisor：Dr. Hiroshi Masuhara

國立交通大學

應用化學系分子科學碩士班

碩士論文

A Thesis

Submitted to M. S. Program

Institute of Molecular Science

Department of Applied Chemistry

National Chiao Tung University

in partial Fulfillment of the Requirements

for the Degree of

Master

in

Molecular Science

July 2011

Hsinchu, Taiwan, Republic of China

(3)

i

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

研究生：劉宗翰指導教授：增原宏博士

國立交通大學

應用化學系分子科學碩士班

摘要

使用雷射誘使結晶化的發生已成為現今取得分子結晶相當重要的技術，

特別是已於 2002 年被成功示範的飛秒雷射誘使結晶化技術，現正往蛋白質

結晶迅速發展。當高強度的雷射脈衝被聚焦於過飽和水溶液之中，水的多

光子吸收現象發生，並對周圍環境產生機械性壓力波，最後致使溶質結晶。

於本研究中，我們探討飛秒雷射參數如脈衝能量、脈衝頻率與聚焦位置對

結晶化與取得結晶的影響，並示範利用聚焦雷射於溶液表面可有效提升分

子結晶的機率。

在第一部分，我們示範飛秒雷射誘發甘胺酸過飽和溶液結晶化並檢驗雷

射參數與結晶機率、結晶形貌和晶相的關係。脈衝能量與頻率的依存性實

驗結果顯示結晶形貌與數量主要是由水的多光子吸收現象引發的空穴氣泡

產生頻率所決定。此外，藉由聚焦雷射脈衝於空氣溶液介面可大幅增進結

(4)

ii

晶機率，其原因被推斷可能為甘胺酸分子於介面的分子吸附，並依此推論

以我們所知首次成功示範使用單發雷射脈衝誘發甘胺酸單晶的產生。

於第二部分，我們以類似上述的方式示範飛秒雷射誘發於蛋白質結晶學

中最標準的酵素與蛋白質-溶菌酶過飽和溶液結晶化。我們發現相較於聚焦

雷射在溶液液滴中與自然結晶的條件下，聚焦雷射在溶液液滴的空氣溶液

介面可顯著地增進結晶機率。另從各條件下取得之結晶的 X-ray 繞射分析結

果可得知其晶格結構並無差異。儘管有研究指出蛋白質會吸附於溶液表面

且會致使蛋白質變性，但實驗結果顯示鹽析效應可穩定蛋白質於溶液介面

的結構，而相關的當代研究亦支持此結果。

目前的研究結果表示於飛秒雷射誘發結晶化中使用雷射照射溶液表面

可改善現存的結晶技術。分子在溶液表面成核的初期階段的進一步光譜動

態研究可望從此結晶方法建立新穎的表面/介面分子科學。

(5)

iii

Femtosecond Laser-Induced Crystallization of

Amino Acid and Protein at the Air/Solution Interface

Student: Tsung-Han Liu Advisor: Dr. Hiroshi Masuhara

M. S. Program

Institute of Molecular Science

Department of Applied Chemistry

National Chiao Tung University

Abstract

Laser-induced crystallization is now becoming an indispensable technique to obtain

molecular crystal. Particularly, femtosecond (fs) laser-induced crystallization was

demonstrated in 2002 and is developing especially toward protein crystallization technique.

When an intense fs laser pulse is focused into a supersaturated solution, multiphoton

absorption of solvent occurs and induces mechanical stress to the surrounding area, which

results in crystallization of solute. In this study, we explore how fs laser parameters such as

pulse energy, repetition rate, and focal position affect crystallization and obtained crystals,

and then demonstrate that solution surface irradiation of fs laser pulses can improve molecular

crystallization probability drastically.

Firstly, fs laser induced crystallization of glycine from the supersaturated solution is

demonstrated and relations among laser parameters and crystallization probability, crystal

(6)

iv

show that the frequency of cavitation bubble generation induced by multiphoton absorption of

water mainly determines crystal morphology and number. Furthermore, significant increase of

crystallization probability is also demonstrated by focusing fs pulses at the air/solution

interface, whose mechanism may be ascribed to molecular adsorption at the interface. Based

on the inference, we have succeeded in single glycine crystal formation induced by single fs

pulse irradiation for the first time as far as we know.

Secondly, we demonstrate fs laser induced crystallization of lysozyme as the most

standard enzyme and protein in protein crystallography in the similar manner to glycine. It is

revealed that irradiation at the air/solution interface of the solution droplet much improved the

crystallization probability, compared with that of fs laser irradiation inside the solution and

that of spontaneous crystallization. X-ray diffraction analysis of the obtained crystals clarified

that crystal structure was the same under each crystallization condition. It is known that

protein can be adsorbed and be localized at the solution surface leading to the denaturation.

However, this result indicates that the salting-out effect can stabilize the structure even at the

surface, which is contemporarily reported in elsewhere.

The present result clearly shows that utilizing solution surface irradiation in the fs

laser-induced crystallization sharpens the existent crystallization technique. Further

spectroscopic dynamics study on the early stage of molecular nucleation occurring at the

surface is promising to establish novel interface/surface molecular science from this

(7)

v

Acknowledgement

Firstly, I would like to sincerely express my appreciation to Prof. Hiroshi Masuhara

(Department of Applied Chemistry and Institute of Molecular Science, National Chiao Tung

University) for supervising my research works during these two years. I feel so fortunate to

have met him and his team, and have a chance to be one of his first students in Taiwan. Prof.

Masuhara is always patient and willing to share opinions with me. I really learned a lot from

his thinking way and attitude.

I deeply appreciate Dr. Takayuki Uwada (Department of Applied Chemistry and Institute

of Molecular Science, National Chiao Tung University) and his direction. Dr. Uwada has

given many lectures related to my research work, and is always willing and eager to assist me

in conducting experiments and give comments on scientific writing. I am much indebted to

him for his great help.

I wish to thank Prof. Atsushi Miura, Dr. Anwar Usman, Dr. Ken-ichi Yuyama

University) and Prof. Teruki Sugiyama (Instrument Technology Research Center, National

Applied Research Laboratories). Prof. Miura takes care of not only laboratory affairs but also

daily lives of students. Dr. Usman helped me a lot from the initial stage of my study about

crystallization. Dr. Yuyama participated in several meeting with students, and accordingly

gave some appropriate comments. Prof. Sugiyama gave countless specific comments and

advices on crystallization.

(8)

vi

(Faculty of Science and Engineering, Kinki University, Japan), Prof. Yaw-Kuen Li

University), Prof. Chun-Jung Chen and Mr. Ting-Wei Jiang (Life Science Group, Scientific

Research Division, National Synchrotron Radiation Research Center and Department of

Physics, National Tsing Hua University). With their indispensible contributions to the related

works, this master thesis becomes more complete and fruitful.

Many thanks to senior members inclusive of Ms. Wen-Yu Lee, Mr. Ping-Yu Hee and Ms.

Jing-Ru Tu, my classmate Mr. Chong-Wei Huang, and junior members inclusive of Mr.

Shun-Fa Wang, Mr. Wei-Yi Chiang, Mr. Tsu-Wei Hsu, Ms. Ling-Ting Huang, Mr. Yan-Hua

Huang and Mr. Ching-Hsu Tseng whose accompany with me and help me a lot in LBNS Lab.

I will certainly cherish the pleasant memories with all the members in LBNS Lab.

Thanks to Prof. Youichiroh Hosokawa, Ms. Kurisu, Mr. Iino (Graduate School of

Materials Science, Nara Institute of Science and Technology, Japan), and former NAIST

members inclusive of Dr. Rungsimanon, Dr. Maezawa, Mr. Ishiguro, Mr. Hiraoka and Ms.

Matsui. We had two workshops and exchanged opinions frequently.

Finally, I would like to express my appreciation to my parents. Although I was not good

at physics in senior high school, I accidentally entered Department of Electrophysics. After I

minored in applied chemistry, they recognized my decision to change major from physics to

physical chemistry from master course and support me until now. In fact, after master course

(9)

vii

List of Figures

Chapter 1 Introduction

Fig. 1.1 Vapor diffusion for protein crystallization. (a) hanging drop method (b) sitting drop

method ... 6

Fig. 1.2 Schematic illustration of crystal formation from a disordered medium to a highly

ordered phase ... 6

Fig. 1.3 Schematic illustration of the formation of ordered crystalline nuclei as a result of the

superposition of a density and a structure fluctuations ... 7

Fig. 1.4 (a) Mechanism of the photochemically induced nucleation of lysozyme (b)

enhancement of protein crystallization by a photochemical reaction ... 8

Fig. 1.5 Diagram showing the ray optics of a spherical Mie particle trapped in water by the

highly convergent light of a single-beam gradient force trap... 10

Chapter 2 Principle

Fig. 2.1 Schematic illustration of multiphoton absorption ... 18

Fig. 2.2 Absorption coefficient spectrum of water, together with the wavelength positions of

the most widely used lasers ... 18

Fig. 2.3 Gibbs free energy versus specific volume of a pure substance for several temperatures

starting at the saturation temperature up to the spinodal temperature ... 19

Fig. 2.4 Variation of vapor bubble nucleation rate with superheat temperature ... 19

Fig. 2.5 Schematic illustration of laser tsunami in water ... 19

Fig. 2.6 Plasma, shock wave, and cavitation bubble produced by Nd:YAG laser pulses of

(13)

xi

Fig. 2.7 Experimentally determined shock wave pressure ... 20

Fig. 2.8 Hydrophone signals measured at a distance of 10 mm from the emission center of the shock waves ... 20

Fig. 2.9 The diameter of the cavitation bubble in water as a function of time ... 21

Fig. 2.10 Spatially limited transient pressure generated by femtosecond laser can induce molecular nucleation in supersaturated solution ... 21

Fig. 2.11 Microscopic images of generated anthracene crystals (a) and a growth process of a bending film-like crystal created at the surface of a large laser-induced bubble (b-d) ... 22

Chapter 3 Experiment Fig. 3.1 Picture of mode-locked Ti:sapphire laser, Tsunami ... 24

Fig. 3.2 Absorption and emission spectra of Ti:sapphire ... 25

Fig. 3.3 The mode-locking principle in Tsunami ... 25

Fig. 3.4 Picture of Ti:sapphire regenerative amplifier system, Spitfire Pro ... 25

Fig. 3.5 The schematic illustration of CPA principle ... 26

Fig. 3.6 Laser light source and microscopic system for crystallization ... 27

Fig. 3.7 Picture of inverted microscope and other attachments ... 27

Fig. 3.8 Reflectance spectrum of the dichroic mirror used in the inverted microscope ... 27

Fig. 3.9 Transmittance spectrum of the objective lens used in the inverted microscope ... 28

Fig. 3.10 Illustration of the du Noüy ring method ... 30

Fig. 3.11 Illustration of the capillary method ... 30

Fig. 3.12 Picture of 1.5 ml glass bottle with 0.5 ml glycine solution inside ... 32

Fig. 3.13 Picture of crystallization plate ... 34

Chapter 4 Glycine Crystallization Fig. 4.1 The chemical structure of glycine ... 35

(14)

xii

Fig. 4.2 The shapes of (a) -polymorph and (b) -polymorph glycine crystals ... 36

Fig. 4.3 Pulse energy (J/pulse) dependence on glycine crystallization probability (%). ... 38

Fig. 4.4 The CCD image of cavitation bubble generation ... 39

Fig. 4.5 (a) Crystallization probability (%) and crystal morphology dependence on repetition

rate (Hz). (b) Images obtained from CCD camera. (c) Picture taken by general camera. ... 40

Fig. 4.6 The crystallization probability (%) depending on the repetition rate (Hz) at the

air/solution interface or glass/solution interface. ... 42

Fig. 4.7 The crystallization probability (%) depending on the distance from the air/solution

interface (mm). ... 42

Fig. 4.8 The crystallization probability (%) depending on the repetition rate (Hz) at 3.0 M, 3.5

M and 4.0M ... 43

Fig. 4.9 The crystals obtained with different exposure times... 45

Fig. 4.10 (a) Raman image of obtained multiple crystals (b) Raman spectrum of obtained

glycine crystal ... 46

Fig. 4.11 Mass spectra of standard sample and laser-induced crystal ... 47

Fig. 4.12 Glycine molecules in one unit cell. (Single crystal case) ... 49

Fig. 4.13 Illustration to explain the morphology difference because of the different repetition

rates ... 50

Chapter 5 Application to Protein

Fig. 5.1 Three-dimensional conformation of hen egg white lysozyme (HEWL) ... 55

Fig. 5.2 The lysozyme crystal shape ... 55

Fig. 5.3 Crystallization probabilities and crystal images of lysozyme under spontaneous

crystallization, irradiation inside the solution and irradiation at the air/solution interface (NaCl

conc.: 10 mg/ml) ... 57

(15)

xiii

irradiation inside the solution and irradiation at the air/solution interface (NaCl conc.: 10, 20,

30 mg/ml) ... 58

Fig. 5.5 The reconstructed lysozyme molecular structures from the X-ray analysis data under different laser conditions (NaCl conc.: 10 mg/ml) ... 64

Fig. 5.6 Electron density distribution of one of -helices under different laser conditions (NaCl conc.: 10 mg/ml)... 64

Fig. 5.7 Electron density distribution of one of disulfide bonds under different laser conditions (NaCl conc.: 10 mg/ml)... 64

Fig. 5.8 Lysozyme concentration (mg/ml) dependence on surface tension ... 65

Fig. 5.9 NaCl concentration (mg/ml) dependence on surface tension ... 65

Fig. 5.10 Conformation changes of protein take place at interfaces... 66

Fig. 5.11 Salting out can keep lysozyme molecules refolding and can localize them even at the air/solution interface ... 67

(16)

xiv

List of Tables

Chapter 4 Glycine Crystallization

Table 4.1 The lattice constants and R-factors of glycine crystals with different

morphologies…...………...49

Chapter 5 Application to Protein

Table 5.1 The lattice constants of lysozyme crystals under various experimental

(17)

1

Chapter 1 Introduction

Laser light is regarded as a monochromatic light source showing high energy density

with high degree of spatial and temporal coherence. Laser beam can be focused to a very tiny

spot, and can be delivered at a large distance with extremely low divergence [1]. Therefore,

laser exhibits high potential and has been applied widely in diverse fields inclusive of physics,

chemistry, material science and so on. Undoubtedly, laser also plays an essential role in

various contemporary and interdisciplinary studies.

As one of laser applications, photoetching, an essential technique for microfabrications,

was first demonstrated when a pulsed ultraviolet excimer laser light with high enough

intensity was irradiated at a surface of organic polymer film and some morphological changes

were induced [2, 3]. Laser ablation was given to describe this phenomenon induced by pulse

laser irradiation at a solid material with intensity higher than a threshold [4]. If laser ablation

takes place in liquid or gaseous phase, the phenomenon is usually described as optical

breakdown. In fact, optical breakdown can be generated in water when focused femtosecond

(18)

2

transition of water from liquid to gas gives transient mechanical stress. The optical breakdown

has been employed for biological applications such as cellular surgery [6, 7], and the

consequent mechanical stress is applied to cellular arrangement or micropatterning [8] and

nanoparticle injection to single cells [9]. Nowadays, the application of optical breakdown has

been expanding to the studies on crystallization and crystal growth and has received much

attention [10-13].

Crystal is a solid material composed of atoms, molecules or even macromolecules which

are three-dimensionally arranged in a long-range order, and crystallization is the process from

atoms or molecules to a crystal. Making crystal is not only an important technique for

purification and separation, but also an indispensable process in determining crystal structures

both of inorganic solid-state materials and biological molecular crystals at atomic resolution

with X-ray diffraction analysis [14].

Among the biological macromolecules inclusive of proteins, nucleic acids and

carbohydrates, proteins constitute the largest group. Enzymes are the most diverse class of

proteins because nearly every chemical reaction in a cell requires a specific enzyme. In order

to understand cellular processes or design new structure-based drugs or pharmacological

(19)

3

and other macromolecules are undoubtedly vital [15, 16]. X-ray diffraction of crystals and

nuclear magnetic resonance (NMR) are two widely-used techniques for the structural

determination of macromolecules at atomic resolution. While NMR does not require protein

crystals and provides more detailed information on the dynamics of the molecule in question,

it can be used only for biopolymers with molecular weight of less than 30,000. X-ray

crystallography can be applied to compound with molecular weight up to at least 106. For

many proteins, the difference is decisive in favor of X-ray diffraction [15]. Accordingly, a

major challenge to the full exploitation of X-ray diffraction technique is that the protein must

first be crystallized [16].

Protein crystallization is mainly a trial-and-error procedure in which protein is slowly

precipitated from its solution, so obtaining suitable single crystals is the least understood step

in the X-ray structural analysis of a protein [15]. As a matter of fact, there are still some

additional requirements for the crystallinity and size of the crystals if precise analysis results

are more likely to be obtained [17]. Moreover, since crystallization is a rate-limiting step from

single molecules to a crystal for some proteins such as membrane proteins which can be

attributed to inherent protein flexibility and to conformational inhomogeneity, it is also

necessary to shorten the crystallization time and elevate the crystallization possibility

(20)

4

techniques have received much attention and have been proposed continually to try to satisfy

the present situation.

1.1 Introduction to conventional crystallization methods

Because amino acids are polar molecules, most of them can be dissolved well in water

rather than organic solvent. Accordingly, water is often used as solvent in crystallization. In an

aqueous solution at pH near isoelectric point, amino acids are in zwitterionic form, and their

zwitterionic property is still kept in crystal. For crystallization, general methods such as

cooling down saturated solution slowly, evaporating saturated solution slowly with adding

polar solvent such as alcohol, adding poor solvent into saturated solution, and adjusting pH to

lower the solubility of the solution are often used. If crystals with high quality are more likely

to be obtained, slow crystal growth at low supersaturated degree is preferred [14].

The history of protein crystal growth can be traced back about more than 150 years ago.

The first observed protein crystallization was published by Hünefeld in 1840, and the protein

was hemoglobin from earthworm. The crystals were obtained when the blood of an

earthworm was pressed between two slides of glass and allowed to dry very slowly. The

above description revealed that protein crystals can be obtained by controlling evaporation of

(21)

5

dehydration which is one of the most acceptable concepts until now. Other conventional

methods such as temperature variation under otherwise constant conditions, the use of salt or

organic solvents as precipitating or crystallizating agents, and the use of metal ions were also

developed in the following decades [19].

The oldest crystallization technique is batch method which has been used for proteins

and nucleic acids crystallization for over 150 years and is regarded as one of the most reliable

crystallization methods. Batch method is attractive and convenient because of its inherent

simplicity and reproducibility. It requires nothing more than direct mixing of an unsaturated

protein solution with a precipitant solution and waiting for a period of time until spontaneous

nucleation commences. For instance, lysozyme, the most widely studied protein for protein

crystallization, is easily crystallized by batch method [19].

The most commonly used technique for protein crystallization is vapor diffusion

inclusive of hanging drop and sitting drop methods. Both of these two methods contain

sample solution and reservoir solution, and require a closed environment. At the early stage,

the sample solution has insufficient precipitate concentration for crystallization. After vapor

diffusion of solvent from sample solution to reservoir solution takes place and vice versa, the

(22)

6

Fig. 1.2 Schematic illustration of crystal formation from a

disordered medium to a highly ordered phase [20]. amounts of solution are quite different between these two [14].

1.2 Introduction to crystallization process

Fig. 1.2 shows a schematic illustration of crystal formation from a disordered medium.

Compared with the final crystal phase, the initial disordered phase has higher free energy.

Crystallization starts from nucleation, in which a tiny embryo of a highly ordered phase with

lower free energy is formed. Nucleation seldom takes place spontaneously at unsaturated

concentration since it is relatively difficult for molecules to overcome some energy barriers

between these two phases. However, at high enough supersaturation, the energy barriers

decrease and the initial state becomes unstable, so that a tiny fluctuation can lead to the

Fig. 1.1 Vapor diffusion for protein crystallization. (a) hanging drop

method (b) sitting drop method

(23)

7

appearance of the new phase [20].

An illustration of phase separation during nucleation in detail is shown in Fig. 1.3. There

are two possible pathways from individual molecules to nucleation. When the size of

molecule accumulation increases to a critical size, the ordered nucleus forms and starts to

form crystal. On the other hand, dense liquid droplet or liquid-liquid phase separation

formation can be observed if the size of molecule accumulation continuously increases [20].

Fig. 1.3 Schematic illustration of the formation of ordered

crystalline nuclei as a result of the superposition of a density and a structure fluctuations [20].

(24)

8

1.3 Light-induced crystallization

Light-induced crystallization can be realized with various light sources through different

mechanisms such as photochemical and nonphotochemical (or photophysical) processes. The

used light source can be Xe lamp, continuous wave laser or pulse laser at different

wavelengths ranged from UV region to NIR region.

1.3.1 Photochemical reaction-induced nucleation

Okutsu et al. have reported photochemical reaction-induced nucleation of anthracene

[21], hen egg-white lysozyme (HEWL) [22], thaumatin [23], and ribonuclease A (RNaseA)

[24]. In the lysozyme crystallization experiment, they exposed UV light (wavelength; 280,

300 and 400 nm) from Xe short arc lamp (wavelength; 200-800 nm, USHIO UXL-300D) to

Fig. 1.4 (a) Mechanism of the photochemically induced

nucleation of lysozyme (b) enhancement of protein crystallization by a photochemical reaction [22].

(25)

9

supersaturated but metastable lysozyme solution and produced photochemical intermediate

(Fig. 1.4). Since the photoproduct has lower solubility in mother solution, it works as a

nucleus and triggers crystallization eventually [22]. Okutsu et al. gave a representative

demonstration of photochemical reaction-induced nucleation. However, it is not known for

sure how radical formation affects the purity of the obtained protein crystal.

1.3.2 Kerr effect-induced nucleation

Garetz et al. discovered that intense nanosecond NIR laser pulses (pulse duration; 20 ns,

repetition rate; 10 Hz, wavelength; 1.06 m) can induce nucleation of urea solution in 1996.

This phenomenon was named as nonphotochemical laser-induced nucleation (NPLIN).

Because it was observed that the initially forming needle-shaped crystals tend to be aligned

parallel to the electric field vector of the light, they proposed that the electric field-induced

realignment of urea molecules results in cluster formation and this mechanism is known as the

optical Kerr effect [25]. Nucleation of glycine [26] was also demonstrated, and the polymorph

of glycine was controlled by changing polarization of laser beam [27] or by additionally

(26)

10

1.3.3 Laser trapping crystallization

Continuous wave (CW) laser-induced crystallization has been developed very fast

recently. Fig. 1.5 shows the principle of laser trapping. When a highly focused laser beam

passes through sphere-like medium with a greater refractive index from surrounding, the

sphere-like medium acts like a weak positive lens and it will be moved toward the focal point

by a substantial net backward trapping force [29]. In laser trapping crystallization, tightly

focused CW NIR laser beam is employed to gather molecules or clusters, forming a high

concentration area at the focal spot, where the nucleation is eventually achieved. Since laser

trapping is actually a dynamic process against diffusion of trapped molecules or clusters from

the focal spot, in order to reach stable laser trapping condition, laser irradiation at a certain

point for a period of time is necessary. Therefore, selecting laser wavelength at which both of

solvent and solute have merely no absorption is quite important. Photons at NIR wavelength

Fig. 1.5 Diagram showing the ray optics of a spherical Mie

particle trapped in water by the highly convergent light of a single-beam gradient force trap [29].

(27)

11

have lower energy and can be less absorbed by biological molecules or systems. However,

water as a common solvent is not applicable since it has vibrational absorption over 1000 nm

[30]. The absorption of water generates a serious problem in temperature elevation which

increases the solubility of solution, and crystallization becomes more difficult. In order to

solve this problem, it is necessary to replace water with deuterated water (heavy water) as

solvent in this crystallization method.

Sugiyama et al. has firstly demonstrated crystallization of glycine by an intense focused

CW-YVO4 laser beam (wavelength; 1064 nm) at the air/solution interface. It was observable

by a CCD camera that the glycine crystal grew from focus to a certain size within a few

seconds [31]. Furthermore, Rungsimanon et al. have investigated the method to control

glycine phase by tuning laser power. It was found that the competing result between photon

pressure and temperature elevation at a certain high laser power led to the probability increase

to find -polymorph glycine crystal, which is not available under ambient conditions [32]. On

the other hand, Yuyama et al. have demonstrated the preparation of a millimeter-scale dense

liquid droplet of glycine induced by laser trapping effect. After focusing a CW NIR laser

beam at the glass/solution interface of a thin film of its supersaturated heavy water solution

and forming the droplet, they found that crystallization starts immediately just after the focal

(28)

12

possibly the early stage of the multistep crystallization process and plays an important role in

photon pressure-induced crystallization of glycine [33].

1.3.4 Molecular assembling induced by optical trapping

Tsuboi et al. reported that the aggregation of lysozyme as nucleus induced by optical

trapping in solution can trigger nucleation [34]. They also observed the assembling of amino

acids (glycine, proline, serine and alanine) induced by laser trapping in solution, and

considered that the observed glycine assembling may be experimentally verified as a

precursor for the crystal nucleus by Sugiyama et al. [35]. The differences between inducing

molecular aggregation as nucleus and forming a local high concentration area by trapping

laser are the focal position and the necessary time for observation. Molecular assembling is

induced in solution and takes several hours for observation, whereas laser trapping

crystallization can only be achieved at the air/solution interface and the crystal grows within

several seconds.

1.3.5 Femtosecond laser-induced crystallization

Femtosecond laser-induced crystallization of HEWL was firstly demonstrated in 2003 by

(29)

13

tosylate) [11], anthracene [13] and glucose isomerase (GI) [36] crystallizations were also

reported in following years. The mechanism will be introduced in detail in Chapter 2.

1.4 Motivation

In this work, we have focused on femtosecond laser-induced crystallization and tried to

clarify how laser parameters inclusive of pulse energy, repetition rate, focal position and

exposure time affect the crystallization probability, crystal morphology and crystalline phase.

Firstly, we demonstrated glycine crystallization since glycine crystal shows several

polymorphs and is well-studied in the field of crystal chemistry [14]. Then we applied our

obtained results to lysozyme crystallization since lysozyme is also a representative protein in

protein crystallography [19]. It is expected to sharpen present crystallization techniques with

optimizing laser parameters and simultaneously establish a remarkable milestone to study

molecular dynamics of femtosecond laser-induced crystallization under appropriate

(30)

14

Chapter 2 Principle

The mechanism of femtosecond laser-induced crystallization is completely different

from other methods because only this technique triggers nucleation by multiphoton absorption

of solvent leading to optical breakdown and is less related with solute. Firstly, principle of

multiphoton absorption is explained with its schematic illustration in Fig. 2.1. Since the

energy of a single photon is not enough to excite a molecule from the ground state to a higher

electronic state, the molecule can be excited only by absorbing more than one photons

simultaneously and the energy difference between the two states is equal to the sum of the

energy of the photons. Figure 2.2 reveals that water has much higher absorption in UV region

due to electronic transition than in visible and NIR region due to vibrational overtone

absorption [37], so it supports that water molecule are excited by photons at 800 nm through

multiphoton absorption.

By focusing intense laser into water, multiphoton absorption of water occurs, and leads

to optical breakdown of water giving vigorous evaporation if the superheated condition is

(31)

15

rapid than the rate of energy consumed by vaporization and normal boiling, the liquid water is

driven to a metastable superheated state. The liquid can remain metastable until the spinodal

temperature is reached. By viewing a plot of the Gibbs free energy versus specific volume at

ambient pressure as shown in Fig. 2.3, unique features of the spinodal temperature can be

appreciated. At the spinodal temperature, the equilibrium between saturated liquid state and

saturated vapor state becomes extremely unstable, so the liquid undergoes spinodal

decomposition, a spontaneous process by which a thermodynamically unstable liquid relaxes

toward equilibrium. The superheated liquid transits to the vapor phase via spinodal

decomposition spontaneously. The generated vapor bubbles grow with increasing temperature,

and eventually stop growing and collapse at the maximum spinodal temperature just below

the critical temperature [38].

Although optical breakdown can be achieved in various common solvents such as water,

deuterated water and some organic solvents, water is still chosen as the best solvent for

crystallization in this study since water is easier to be obtained with much less cost, is more

environmentally friendly and always plays an important role in biological systems. Different

from laser trapping crystallization method in which deuterated water has been used as solvent

to prevent from temperature elevation induced by continual 1064 nm CW laser irradiation, it

(32)

16

crystallization. Moreover, the wavelength of femtosecond laser can be adjusted to 800 nm at

which it is well-known that neither water nor biological molecules such as amino acids and

proteins have strong vibrational overtone absorption. Therefore, femtosecond laser is regarded

as an appropriate light source to deal with biological molecules or systems such as proteins,

cells and tissues which usually contains large amount of water and has been widely applied

[30].

When an intense femtosecond laser light is focused into water, multiphoton absorption of

water induces nonlinear phenomena such as shockwaves, cavitation bubbles, and jet flow

formations in water (Fig. 2.5) [5]. These nonlinear phenomena were also described as laser

tsunami. A real image of optical breakdown of water was shown in Fig. 2.6 while it was

induced by picosecond laser irradiation but not femtosecond laser irradiation. Trace of shock

wave propagation and cavitation bubble generation can be clearly observed, and the spatial

and temporal intensity decays of local transient pressure induced by shock wave were also

recorded as shown in Fig. 2.7 and Fig. 2.8, respectively [39].

Lord Rayleigh discussed the collapse behavior of a spherical cavity in a homogeneous

incompressible liquid and established a mathematical model to calculate time-dependent radii

(33)

17

cavitation bubbles generated by femtosecond laser pulses, and compared the results with

theoretical calculations derived by L. Rayleigh (Fig. 2.9) [41].

Laser tsunami as local physical perturbations in supersaturated solutions has been

utilized to trigger the nucleation by increasing local concentration as shown in Fig. 2.10

[10-13, 36]. Since the duration time of femtosecond laser pulse is too short to accumulate heat

near focal point, it is considered that femtosecond laser-induced crystallization takes place

under constant temperature.

Recently, high-speed and high-resolution photography have also been used to investigate

laser-induced cavitation bubble behavior and the subsequent nucleation of crystals. The

bubble size expanding and shrinking periodically is called Rayleigh bubble. In order to

discuss the importance of Rayleigh bubble to crystallization, Murai et al. increased the

viscosity of lysozyme solution by adding agarose gel against protein diffusion from the focal

point and recorded the conformation change of bubbles. They successfully observed the

formation of locally high concentration region induced by bubble conformation change

leading to elevate crystallization probability around the surface of bubble [42, 43]. Nakamura

et al. previously observed film-like crystal formation of anthrathene along with the bubble

(34)

18

Another supporting evidence is that Soare et al. observed ring-shaped optical disturbance

which originates from refractive index change induced by crystal nuclei formation after

collapse of cavitation bubble [44].

Fig. 2.1 Schematic illustration of

multiphoton absorption.

S0

S1

Fig. 2.2 Absorption coefficient spectrum of water, together with the

(35)

19

Shock waves _{Cavitation bubbles} Focusing femtosecond laser

Fig. 2.5 Schematic illustration of laser tsunami

in water.

Fig. 2.3 Gibbs free energy versus specific volume of a pure substance for several

temperatures starting at the saturation temperature, Tsat, where the liquid and gaseous

states are in equilibrium, up to the spinodal temperature, Tspin. Note that the local

minimum corresponding to the liquid phase disappears at the spinodal temperature, effectively forcing conversion of the substance to the vapor phase [37].

(36)

20

Fig. 2.7 Experimentally determined shock wave pressure after pulses of 30 ps duration

with 50 J and 1 mJ energy. The numbers indicate the local slope of the curves [38].

Fig. 2.8 Hydrophone signals measured at a distance of 10 mm from the emission

center of the shock waves. The respective values of laser pulse duration and energy are (a) 30 ps, 50 J; (b) 30 ps, 1 mJ [38].

Fig. 2.6 Plasma, shock wave, and cavitation bubble produced by Nd:YAG laser pulses

of different duration and energy: (a) 30 ps, 50 J; (b) 30 ps,1 mJ. All pictures were taken 44 ns after the optical breakdown [38].

(37)

21

Fig. 2.9 The diameter of the cavitation bubble in water as a

function of time. The laser fluence is 9 J/cm2 [40].

Fig. 2.10 Spatially limited transient pressure generated by

femtosecond laser can induce molecular nucleation in supersaturated solution.

(38)

22

Fig. 2.11 Microscopic images of generated anthracene crystals at a pulse energy of 6.7

J/pulse (a) and a growth process of a bending film-like crystal created at the surface of a large laser-induced bubble at a pulse energy of 16.5 J/pulse (b-d) [13].

(39)

23

Chapter 3 Experiment

3.1 Experimental setup

3.1.1 Femtosecond laser light source

Amplified Ti:sapphire femtosecond laser which is composed of a mode-locked

Ti:sapphire laser and Chirp Pulse Amplification (CPA) was used as the light source in this

study. A mode-locked Ti:sapphire laser light (wavelength; 700-900 nm, repetition rate; 80

MHz, pulse duration; 160 fs, Tsunami, Spectra Physics) was generated when it was pumped

by a continuous wave solid-state visible laser (wavelength; 532 nm, Millenia Pro, Spectra

Physics) at about 7 W and was controlled by an electronics module (Model 3955, Spectra

Physics). A prism sequence and a slit were used for dispersion control and wavelength

selection, respectively. Spectral information of the output laser light was provided by a fiber

optic spectrometer (USB4000, Ocean Optics). After full width at half maximum (FWHM) of

the spectrum and central wavelength of the laser light were determined to be about 12 nm and

800 nm simultaneously, the laser light was amplified by a Ti:sapphire regenerative amplifier

(40)

24

Since self-focusing, a nonlinear optical effect in which an intense light beam modifies

the refractive index of the material it is passing through, takes place and causes permanent

damage to the Ti:sapphire crystal, it is necessary to limit the peak power of a pulse below 10

GW/cm2 during amplification. Therefore, CPA was applied to the system to amplify pulse

beyond the peak power and to keep the power density far below the damage threshold of the

crystal. CPA was accomplished in the following three steps (Fig. 3.5). The first step stretched

the seed pulse supplied from a stable and mode-locked femtosecond laser to about 104 times.

Stretching the pulse reduces its peak energy, which greatly reduces the probability of damage

to the crystal at the same time. The second step amplified the stretched pulse by passing it

through the Ti:sapphire regenerative amplifier. Finally, a set of dispersive optics was used to

recompress the stretched, amplified pulse as close as possible to its original duration and high

peak power compressed pulse was generated.

Fig. 3.1 Picture of mode-locked Ti:sapphire laser,

(41)

25

Fig. 3.2 Absorption and emission spectra of

Ti:sapphire.

Fig. 3.3 The mode-locking principle in Tsunami.

Fig. 3.4 Picture of Ti:sapphire regenerative amplifier

(42)

26

3.1.2 Experimental setup

The experimental setup is shown in Fig. 3.6. Linearly polarized femtosecond laser pulses

from femtosecond laser (wavelength; 800 nm, pulse duration; 160 fs, Spitfire Pro, Spectra

Physics) were introduced to an inverted microscope (IX-71, Olympus) through an objective

lens (10×, N.A. 0.25, PlanN, Olympus). Pulse energy was adjusted by using a half-wave plate,

a polarizing beam splitter, and a variable neutral density filter, and was measured after an

objective lens by energy meter (842-PE, Spectra Physics). The repetition rate of femtosecond

laser pulse train was controlled by a Pockels Cell. The crystal formation and crystal

morphology were observed with a digital CCD camera (CV-S3200N, JAI) attached to the

microscope. All the experiments were carried out at room temperature (22 ̊C). [45]

(43)

27

Fig. 3.6 Laser light source and microscopic system for crystallization.

Fig. 3.7 Picture of inverted microscope and other attachments.

Fig. 3.8 Reflectance spectrum of the dichroic mirror used in the inverted microscope

(44)

28

3.1.3 Confocal Raman microspectroscopic system

Confocal Raman microspectroscopic measurement was performed to characterize the

crystalline phase of obtained crystals with a confocal unit coupled optical inverted microscope

(FV300, Olympus). Raman signals were obtained by exposing 532 nm DPSS laser beam (40

mW, JLW-532-200, SLOC) through a line filter and an objective lens (10×, N.A. 0.25, PlanN,

Olympus) to the crystals placed on the microscope stage, and were recorded by a CCD

camera combined with polychromator with a 150 gr/mm grating (polychromator; SpectraPro

2300i, CCD camera; PIXIS400, Princeton instruments) after passing through a confocal

aperture (500 m). By fast laser scanning a certain detecting two-dimensional area of the

obtained crystals with a Galvano mirror unit, Raman images were also recorded.

Fig. 3.9 Transmittance spectrum of the objective lens used in the inverted

(45)

29

3.1.4 Surface tension measurement

Surface tension is a property of a liquid surface that allows it to resist an external force.

In the bulk of the liquid, each molecule is pulled equally in every direction by neighboring

liquid molecules, resulting in a net force of zero. The molecules at the surface do not have

other molecules on all sides of them and therefore are pulled inwards. This creates some

internal pressure and forces liquid surfaces to contract to the minimal area. Similar to surface

tension, molecular adsorption is related to surface energy. In a bulk material, all the bonding

requirements of the constituent molecules of the material are filled. The molecules on the

surface experience a bonding deficiency because of the dissymmetric surrounding. Therefore,

it is energetically favorable for them to attract adsorbates.

The du Noüy ring method, a relatively simple method with much higher precision

proposed by the French physicist Pierre Lecomte du Noüy in 1925 [46], is applied to measure

surface tension of sample solution here. As Fig. 3.10 shows, a hanging and relatively

immobile thin platinum ring (diameter of wire 2r; 0.3 mm, diameter of ring 2R; 13.5 mm,

Narika) connected with an electronic balance (minimum weight; 0.1 mg, CP124S, Sartorius)

from the bottom is submersed in a solution on a stage, and then the height of the stage is

slowly decreased at a constant speed. As the ring is almost pulled out of the liquid, the

(46)

30

the liquid. The following is the formula:

γ = ………....………(3.1)

W. D. Harkins and H. F. Jordan considered that the shape of liquid film is strongly

dependent on the radius of the ring, the radius of the wire, the density of the solution and the

surface tension, compared with the maximum force F obtained by both the ring method and

the capillary method (Fig. 3.11 [47]), and suggested that the obtained surface tension  with

ring method should be modified by a correction factor (R3/V, R/r) [48]. Therefore, the real

surface tension is shown as the following formula:

γ = , ………(3.2)

Fig. 3.10 Illustration of the du Noüy ring

method

.

Fig. 3.11 Illustration of the capillary

(47)

31

3.1.5 X-ray diffraction

X-rays as electromagnetic waves are diffracted by a crystal in many directions.

Following Bragg’s law, constructive interference of X-rays scattered by a crystal occurs in a

few specific directions. These diffraction data inclusive of intensities and phases in reciprocal

space can derive the dimensions of the unit cell and the molecular structure in real space

through Fourier transform. In fact, X-rays are mainly scattered by the electrons but not nuclei

in the atoms. Therefore, a primitive data of electron density distribution is firstly obtained

from the diffraction pattern. After repeating fitting the atomic model and refinement again and

again, molecular structure at atomic scale can be reconstructed [15].

Particle accelerators as synchrotrons and storage rings are the most powerful X-ray

sources, and they are profitably used by protein X-ray crystallographer for data collection on

weakly diffracting specimens such as very tiny crystals or crystals with extremely large unit

cells. Synchrotrons are facility for circulating electrically charged particles at nearly the speed

of light. The particles are injected into the storage ring directly from a linear accelerator or

through a booster synchrotron. When the particle beam changes direction, the electrons or

positrons are accelerated toward the center of the ring, emit electromagnetic radiation, and

consequently lose energy. This energy loss is compensated by a radiofrequency input at each

(48)

32

which is two orders of magnitude stronger than a conventional X-ray tube. Another advantage

is the low divergence of the beam resulting in sharper diffraction spots [15].

3.2 Sample preparation

3.2.1 Supersaturated glycine aqueous solution

Glycine (99.0%, Lot. No. LTF4344, Wako Pure Chemical) in the form of white powder

was used without further purification. Deionized water as solvent was filtrated with a syringe

filter (pore size; 0.22 m, SLGV 013 SL, Millipore) before dissolving glycine. 1.5 ml glass

bottle (Nichiden Rika Glass) (Fig. 3.12) as container was cleaned by plasma cleaner (ATTO,

diener electronic). 0.5 ml of supersaturated glycine aqueous solution was prepared at 3.0 M,

3.5 M and 4.0 M (Csat= 3.0 M [49]) in glass bottle with cap screwed on. The solution was

slowly heated up from room temperature to 60 ̊C in water bath at heating rate 5 ̊C per hour

until glycine was completely dissolved, and was cooled down to room temperature

spontaneously.

Fig. 3.12 Picture of 1.5 ml glass bottle

(49)

33

3.2.2 Supersaturated lysozyme aqueous solution

3.2.2.1 Acetic acid-sodium acetate buffer aqueous solution

Filtrated deionized water was also used as solvent here. 200 mM acetic acid solution was

prepared by adding 120 l acetic acid (99.8%, Sigma-Aldrich) into 10 ml water. On the other

hand, 200 mM sodium acetate solution was prepared by adding 0.164 g sodium acetate

(Sigma-Aldrich) into 10 ml water. 2.9 ml acidic solution and 2.1 ml basic solution are mixed

and diluted with 5 ml water. Finally, the obtained solution was 100 mM acetic acid-sodium

acetate aqueous solution used as buffer solution and its pH was about 4.5 measured by pH

meter (pHi 510, Beckman Coulter) [50].

3.2.2.2 Supersaturated lysozyme aqueous solution

Lysozyme from egg white or hen egg white lysozyme (HEWL) (for Biochemistry, Lot.

No. PEK3768, Wako Pure Chemical) in the form of white powder was used without further

purification. Supersaturated lysozyme solution was prepared with buffer solution at 40 mg/ml

at room temperature. After lysozyme was completely dissolved in buffer solution, sodium

chloride (>99.9%, FSA Laboratory Supplies) as a crystallization agent was added to slightly

decrease protein solubility. The pH of prepared solution was not much changed after

(50)

34

crystallized successfully at pH around 4.5 [19, 51]. The solution was dropped into wells

fabricated on a plastic crystallization plate (VDX48 Plate with sealant, Hampton) (Fig. 3.13)

with 50 l in each well sealed with glass coverslip (Matsunami).

(51)

35

Chapter 4 Glycine Crystallization

4.1 Introduction

It is well-known that protein is a biological macromolecule composed of 20 common

amino acids. Glycine, NH2CH2COOH, is a common amino acid with lowest molecular weight

and simplest structure (Fig. 4.1), and has been repeatedly studied in the field of crystal

chemistry.

Glycine crystal shows three polymorphs inclusive of two monoclinic with

centrosymmetric space groups (, s.gr. P21/n, and , s.gr. P21) and one trigonal with

non-centrosymmetric space group (, s.gr. P31) [52-54]. The shape difference between - and

-polymorph glycine crystals can be clearly distinguished in Fig. 4.2. Many groups have

employed glycine as a standard sample to laser-induced crystallization methods. It can be

(52)

36

expected that the detailed study on femtosecond laser-induced glycine crystallization will be

more helpful to understand the laser parameters effect by comparison with such previous

work.

Usually, commercially available glycine powder contains mixture of - and

-polymorphs. The polymorph content obtained from different chemical companies or even

just different packages of the same company ranges from  with admixture of  to  with

admixture of . -polymorph crystal is the most thermodynamically stable among these three

polymorphs, but spontaneous crystallization gives -polymorph under almost every

experimental condition. Crystal of -polymorph is preferentially obtained under relatively

severe conditions such as using acid and base water solutions, using heavy water solution, and

adding compounds that inhibit the growth of -glycine like hexafluoravaline. However, based

on our experimental result, it is still difficult to prepare -polymorph crystal under using

Fig. 4.2 The shapes of (a) -polymorph and (b)

(53)

37

heavy water as solvent. In fact, the success of crystallizing -polymorph seems to depend

much more on the presence of the seeds of this polymorph in the powder used for preparing a

saturated solution, than on the solvent used. To obtain -polymorph crystal, ethanol should be

used to precipitate this form from a saturated water solution. However, a mixture of the -

and the -polymorphs can be always obtained [52, 54].

The transformations among these three polymorphs are very complicated [52]. Although

-polymorph is quite stable at room temperature, it transforms to -polymorph after several

months. The transformation rate is highly dependent on the humidity of environment. High

humidity accelerates the transformation rate. -polymorph prefers to transform to - but not

-polymorph under high humidity but can be preserved well in dried environment.

-polymorph also transforms to -polymorph with heating below 100 ̊C or mechanical shocks

such as crush. -polymorph is quite stable at room temperature, but it transforms to

-polymorph at 165 ̊C. And reverse transformation does not take place immediately even after

-polymorph is cooled down to room temperature [14, 53].

4.2 Results

The possibilities of crystallization were recorded as crystallization probability here. The

(54)

38

total number of samples done at the same condition. Experiment was repeated 10 times under

each condition. In order to confirm that the crystals are obtained by femtosecond laser

irradiation rather than spontaneous crystallization, several blank samples under each condition

were prepared for reference simultaneously.

4.2.1 Pulse energy dependence

Pulse energy dependence of glycine crystallization probability is shown in Fig. 4.3. Laser

pulse train was focused at the solution/glass interface, which was actually about several tens

Fig. 4.3 Pulse energy (J/pulse) dependence on glycine crystallization probability

(%). Repetition rate, exposure time, sample concentration and focal position are 1 kHz, 10 minutes, 4.0 M and several tens m from the bottom of the bottle, respectively.

(55)

39

m above the interface, to avoid laser ablation of glass. Cavitation bubble generation

threshold is confirmed by a CCD camera (Fig. 4.4). With increasing the pulse energy,

crystallization probability increased, while it showed a threshold of crystallization around 2.5

J/pulse, which was in good agreement with that of cavitation bubble generation.

4.2.2 Repetition rate dependence

Crystallization probability and crystal morphology dependence on repetition rate of

irradiated femtosecond laser pulse train are recorded in Fig. 4.5. Pulse energy, irradiation time,

and focal position were fixed to 100 J per pulse, 10 minutes and at the solution/glass

interface, respectively. After waiting for several hours from femtosecond laser pulses

irradiation, we successfully obtained glycine crystals in the glass bottle.

It is found that the crystallization probability increased with the repetition rate and it

Fig. 4.4 The CCD image of cavitation bubble generation.

When the pulse energy is 2.5 J per pulse, cavitation bubbles are generated around the focal point.

(56)

40

Fig. 4.5 (a) Crystallization probability (%) and crystal morphology dependence on

repetition rate (Hz). (b) Images obtained from CCD camera. (c) Picture taken by general camera. Pulse energy, exposure time, sample concentration and focal position are 100 J per pulse, 10 minutes, 4.0 M and several tens m from the bottom of the bottle, respectively.

Single crystal Multiple crystal Powder-like crystal (c) (b) (a)

became saturated to 100 % above 250 Hz. On the contrary, at low repetition rate less than 5

Hz, crystals could not be obtained. In addition, it was confirmed that crystal morphology

exhibited a strong correlation with the repetition rate (Fig. 4.5 (b) and (c)). At lower repetition

rates, transparent single crystals with prismatic morphology floating on the solution were

obtained frequently. The size of the single crystals was roughly 5×4×3 mm3. Besides,

powder-like crystals were more likely to be observed at higher repetition rates. These

powder-like crystals are quite many, small, and uniform sized (mean size; 50×20×10 m3).

(57)

41

middle rates around 50 Hz.

It is also found that crystal formation period was closely related to crystal morphology.

In this experiment, single crystal formation took at least 3 to 4 days to be observed. However,

the formation period was dramatically shortened as the size of crystal decreased. Powder-like

crystal formation could be observed during pulse irradiation.

4.2.3 Focal position dependence

The result of focal position dependence is shown in Fig. 4.6. Pulse energy, irradiation

time, and focal position were fixed to 100 J per pulse, 10 minutes and at the air/solution

interface, respectively. Compared with the result of repetition rate dependence at the

solution/glass interface, it was observed that the crystallization probability was significantly

improved even at relatively low repetition rates such as 5 Hz and powder-like crystals were

always obtained during laser irradiation when we irradiated the laser light at the air/solution

interface.

In order to confirm the interface effect, the focal position was shifted from the

solution/glass interface to the air/solution one with 1 mm step (Fig. 4.7). The solution

(58)

42

focal positions with 1 mm interval. As the result, it is found that the crystallization probability

is improved upon both air/solution and glass/solution interfaces irradiation whereas the

probability is higher at the air/solution irradiation. In contrast to their interface irradiation, the

Fig. 4.7 The crystallization probability (%) depending on the distance from the

air/solution interface (mm). Repetition rate, pulse energy, exposure time and sample concentration are 25 Hz, 100 J per pulse, 10 minutes and 4.0 M, respectively.

Air/solution interface

Solution/glass interface

Fig. 4.6 The crystallization probability (%) depending on the repetition rate (Hz) at

the air/solution interface or glass/solution interface. Pulse energy, exposure time and sample concentration are 100 J per pulse, 10 minutes and 4.0 M, respectively.

(59)

43

irradiation inside the solution suppressed the probability in spite of relatively higher

probability at the position closer to air/solution interface. Therefore, we conclude that

utilizing interface can improve crystallization probability in femtosecond laser induced

crystallization and the irradiation at air/solution interface is more efficient.

4.2.4 Concentration dependence

From the view point of crystallography, it is well-known that the quality of the crystal in

low supersaturated solution is much higher than that in highly supersaturated solution, which

has been recognized in the conventional crystallization methods [31]. Therefore, it is expected

to obtain crystals at relatively lower saturated concentration.

Fig. 4.8 The crystallization probability (%) depending on the repetition rate (Hz) at

3.0 M, 3.5 M and 4.0M. Repetition rate and exposure time are 50 J per pulse and 10 minutes, respectively.

(60)

44

Glycine solutions at lower supersaturated concentration and even saturated concentration

(Csat=3.0 M) were irradiated by laser focused at the air/solution interface and the

solution/glass interface. Concentration dependence at three different repetition rates is

presented in Fig. 4.8. It is found that it is possible to get crystals even at lower concentration

(3.5 M) when we irradiated laser pulses at the air/solution interface at 50 and 100 Hz, and the

crystallization probability is apparently higher. However, it is difficult to form crystal when

we irradiated laser pulses at the solution/glass interface at the same repetition rates. Compared

with the case at saturated concentration (3.0 M), it is still difficult to crystallize at neither the

air/solution interface nor the solution/glass interface at even higher repetition rate such as 100

Hz.

4.2.5 Exposure time dependence

Since the crystallization probability reached such a high enough value when we

irradiated laser pulse at the air/solution interface, we tried to reduce the laser irradiation time.

It was expected to simplify the conditions and finally achieve single crystal formation by

single laser pulse, which is the simplest and the most ideal condition to study the molecular

dynamics of femtosecond laser-induced crystallization.

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

國

立

交

通

大

學

應用化學系分子科學碩士班

碩

士 論 文

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

Femtosecond Laser-Induced Crystallization of Amino Acid

and Protein at the Air/Solution Interface

研 究 生：劉 宗翰 (Tsung-Han Liu)

指導教授：增原 宏 博士 (Dr. Hiroshi Masuhara)

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化

Femtosecond Laser-Induced Crystallization of Amino Acid

and Protein at the Air/Solution Interface

研 究 生：劉宗翰

Student：Tsung-Han Liu

指導教授：增原宏 博士

Advisor：Dr. Hiroshi Masuhara

國立交通大學

應用化學系分子科學碩士班

碩士論文

A Thesis

Submitted to M. S. Program

Institute of Molecular Science

Department of Applied Chemistry

National Chiao Tung University

in partial Fulfillment of the Requirements

for the Degree of

Master

in

Molecular Science

July 2011

Hsinchu, Taiwan, Republic of China

飛秒雷射誘發胺基酸與蛋白質於氣液介面結晶化之研究

研究生：劉宗翰 指導教授：增原宏 博士

國立交通大學

應用化學系分子科學碩士班

摘要

使用雷射誘使結晶化的發生已成為現今取得分子結晶相當重要的技術，

特別是已於 2002 年被成功示範的飛秒雷射誘使結晶化技術，現正往蛋白質

結晶迅速發展。當高強度的雷射脈衝被聚焦於過飽和水溶液之中，水的多

光子吸收現象發生，並對周圍環境產生機械性壓力波，最後致使溶質結晶。

於本研究中，我們探討飛秒雷射參數如脈衝能量、脈衝頻率與聚焦位置對

結晶化與取得結晶的影響，並示範利用聚焦雷射於溶液表面可有效提升分

子結晶的機率。

在第一部分，我們示範飛秒雷射誘發甘胺酸過飽和溶液結晶化並檢驗雷

射參數與結晶機率、結晶形貌和晶相的關係。脈衝能量與頻率的依存性實

驗結果顯示結晶形貌與數量主要是由水的多光子吸收現象引發的空穴氣泡

產生頻率所決定。此外，藉由聚焦雷射脈衝於空氣溶液介面可大幅增進結

晶機率，其原因被推斷可能為甘胺酸分子於介面的分子吸附，並依此推論

以我們所知首次成功示範使用單發雷射脈衝誘發甘胺酸單晶的產生。

於第二部分，我們以類似上述的方式示範飛秒雷射誘發於蛋白質結晶學

中最標準的酵素與蛋白質-溶菌酶過飽和溶液結晶化。我們發現相較於聚焦

雷射在溶液液滴中與自然結晶的條件下，聚焦雷射在溶液液滴的空氣溶液

介面可顯著地增進結晶機率。另從各條件下取得之結晶的 X-ray 繞射分析結

果可得知其晶格結構並無差異。儘管有研究指出蛋白質會吸附於溶液表面

且會致使蛋白質變性，但實驗結果顯示鹽析效應可穩定蛋白質於溶液介面

的結構，而相關的當代研究亦支持此結果。

目前的研究結果表示於飛秒雷射誘發結晶化中使用雷射照射溶液表面

可改善現存的結晶技術。分子在溶液表面成核的初期階段的進一步光譜動

態研究可望從此結晶方法建立新穎的表面/介面分子科學。

Femtosecond Laser-Induced Crystallization of

Amino Acid and Protein at the Air/Solution Interface

Student: Tsung-Han Liu Advisor: Dr. Hiroshi Masuhara

M. S. Program

Institute of Molecular Science

Department of Applied Chemistry

National Chiao Tung University

Abstract

Acknowledgement

Table of Contents

List of Figures

List of Tables

Chapter 1

Introduction

1.1 Introduction to conventional crystallization methods

士論文

研究生：劉宗翰 (Tsung-Han Liu)

指導教授：增原宏博士 (Dr. Hiroshi Masuhara)

研究生：劉宗翰

指導教授：增原宏博士

研究生：劉宗翰指導教授：增原宏博士