Influence of UV light and heat on the ferroelectric properties of lithium niobate crystals [Elektronische Ressource] / Hendrik Steigerwald. Mathematisch-Naturwissenschaftliche Fakultät

Hendrik Steigerwald

Description

INFLUENCE OF UV LIGHT AND HEAT ONTHE FERROELECTRIC PROPERTIES OFLITHIUM NIOBATE CRYSTALSDissertationzurErlangung des Doktorgrades (Dr. rer. nat.)derMathematisch-Naturwissenschaftlichen Fakultat¨derRheinischen Friedrich-Wilhelms-Universitat¨ Bonnvorgelegt vonHendrik SteigerwaldausNeuwied am RheinBonn 2011AngefertigtmitGenehmigungderMathematisch-NaturwissenschaftlichenFakultat¨ derRheinischenFriedrich-Wilhelms-Universitat¨ Bonn1. Gutachter: Prof. Dr. KarstenBuse2. Prof. Dr. KarlMaierTagderPromotion: 19.5.2011Erscheinungsjahr: 2011Contents1 Introduction 12 Fundamentals 32.1 Lithium niobate crystals – general properties . . . . . . . . . 32.1.1 Crystal structure and symmetry . . . . . . . . . . . . 32.1.2 Characteristic absorption . . . . . . . . . . . . . . . . 52.1.3 Stoichiometry, defects and doping . . . . . . . . . . . 52.2 Lithium niobate as a ferroelectric material . . . . . . . . . . . 72.2.1 Domain inversion . . . . . . . . . . . . . . . . . . . . . 72.2.2 Coercive field reduction . . . . . . . . . . . . . . . . . 102.2.3 Domain patterning . . . . . . . . . . . . . . . . . . . . 102.2.4 visualization . . . . . . . . . . . . . . . . . . 113 Experimentalmethods 133.1 Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133.2 Experimental setup for domain inversion . . . . . . . . . . . 143.2.1 Sample holder . . . . . . . . . . . . . . . . . . . . . . . 143.2.2 Domain patterning with structured electrodes . . . . 153.

Sujets

• Physik

Informations

• Publié par: rheinische_friedrich-wilhelms-universitat_bonn
• Publié le: 01 janvier 2011
• Nombre de lectures: 6
• Langue: English
• Poids de l'ouvrage: 3 Mo

Extrait

I NFLUENCE OF UV LIGHT AND HEAT ON THEFERROELECTRICPROPERTIESOF LITHIUM NIOBATE CRYSTALS

Dissertation zur Erlangung des Doktorgrades (Dr. rer. nat.)

der Mathematisch-NaturwissenschaftlichenFakultat der RheinischenFriedrich-Wilhelms-UniversitatBonn

vorgelegt von Hendrik Steigerwald

aus Neuwied am Rhein

Bonn 2011

Angefertigt mit Genehmigung der Mathematisch-Naturwissenschaftlichen Fakult¨atderRheinischenFriedrich-Wilhelms-Universita¨tBonn

1. Gutachter: Prof. Dr. Karsten Buse 2. Gutachter: Prof. Dr. Karl Maier Tag der Promotion: 19.5.2011 Erscheinungsjahr: 2011

Contents

1

2

3

Introduction

Fundamentals 2.1 Lithium niobate crystals  general properties . . . . . . . . . 2.1.1 Crystal structure and symmetry . . . . . . . . . . . . 2.1.2 Characteristic absorption . . . . . . . . . . . . . . . . 2.1.3 Stoichiometry, defects and doping . . . . . . . . . . . 2.2 Lithium niobate as a ferroelectric material . . . . . . . . . . . 2.2.1 Domain inversion . . . . . . . . . . . . . . . . . . . . . 2.2.2 Coercive field reduction . . . . . . . . . . . . . . . . . 2.2.3 Domain patterning . . . . . . . . . . . . . . . . . . . . 2.2.4 Domain visualization . . . . . . . . . . . . . . . . . .

Experimental methods 3.1 Samples . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3.2 Experimental setup for domain inversion . . . . . . . . . . . 3.2.1 Sample holder . . . . . . . . . . . . . . . . . . . . . . . 3.2.2 Domain patterning with structured electrodes . . . . 3.2.3 Heating of the sample . . . . . . . . . . . . . . . . . . 3.2.4 Homogeneous UV illumination of the sample . . . . 3.3 Domain inversion and domain imaging  measurement of the coercive field strength . . . . . . . . . . . . . . . . . . . . 3.3.1 Inversion of ferroelectric domains . . . . . . . . . . . 3.3.2 Poling current . . . . . . . . . . . . . . . . . . . . . . . 3.3.3 In-situ visualization . . . . . . . . . . . . . . . . . . . 3.3.4 Domain-selective etching . . . . . . . . . . . . . . . . 3.3.5 Imaging and generation of domains via piezoresponse force microscopy . . . . . . . . . . . . . . . . . . . . .

1

3 3 3 5 5 7 7 10 10 11

13 13 14 14 15 17 17

18 19 19 20 20

22

i

C ONTENTS

4

5

ii

3.4

Domain patterning by local irradiation with strongly ab- sorbed UV light . . . . . . . . . . . . . . . . . . . . . . . . . . 3.4.1 Irradiation setup . . . . . . . . . . . . . . . . . . . . . 3.4.2 Generation of the latent state . . . . . . . . . . . . . . 3.4.3 Poling inhibition . . . . . . . . . . . . . . . . . . . . . 3.4.4 Persistence of the latent state . . . . . . . . . . . . . . 3.4.5 Irradiation of the non-polar faces . . . . . . . . . . . .

Experimental results 4.1 Domain patterning with structured electrodes and ultravi- olet light illumination . . . . . . . . . . . . . . . . . . . . . . . 4.1.1 Domain patterning in Mg-doped congruent lithium niobate . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.1.2 Domain patterning in Mg-doped near-stoichiometric lithium niobate . . . . . . . . . . . . . . . . . . . . . . 4.2 Domain patterning by UV irradiation . . . . . . . . . . . . . 4.2.1 Poling inhibition in near-stoichiometric lithium nio- bate . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.2 Persistence of the latent state . . . . . . . . . . . . . . 4.2.3 Mapping of the coercive field . . . . . . . . . . . . . . 4.2.4 Bulk domain patterning by poling inhibition . . . . . 4.2.5 PI domain patterning for whispering gallery mode resonators . . . . . . . . . . . . . . . . . . . . . . . . . 4.2.6 Direct domain writing on the non-polar faces . . . . . 4.3 Influence of heat and UV light on the coercive field . . . . . 4.3.1 Temperature dependence of the coercive field . . . . 4.3.2 Coercive field reduction by UV illumination at ele- vated temperatures . . . . . . . . . . . . . . . . . . . . 4.3.3 Coercive field reduction of chemically reduced lithium niobate . . . . . . . . . . . . . . . . . . . . . . . . . . .

Discussion 5.1 Domain patterning with structured electrodes . . . . . . . . 5.2 Domain patterning by UV irradiation . . . . . . . . . . . . . 5.2.1 The origin of the latent state . . . . . . . . . . . . . . . 5.2.2 Modeling of lithium thermodiffusion . . . . . . . . . 5.2.3 Discussion of the experimental results in the frame- work of the model . . . . . . . . . . . . . . . . . . . . 5.2.4 Bulk domain patterning by poling inhibition . . . . . 5.2.5 Domain patterning for whispering gallery mode res- onators . . . . . . . . . . . . . . . . . . . . . . . . . . .

23 23 25 25 26 26

27

27

27

29 30

31 32 33 36

37 39 41 41

44

45

47 47 49 49 49

54 56

56

6

5.3

5.4

C STENTON

5.2.6 Direct domain writing on the non-polar faces . . . . . Influence of heat and UV light on the coercive field . . . . . 5.3.1 Influence of heat on the coercive field . . . . . . . . . 5.3.2 Influence of UV light on the coercive field . . . . . . . Outlook . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

Summary

Bibliography

57 62 62 64 64

67

69

iii

C STNETNO

iv

Chapter

1

Introduction

Allwissend bin ich nicht; doch viel ist mir bewusst. [1]

Since the advent of the laser [2], optical technologies have become a part of everyday-life. Applications in fields such as life science [3], medi- cine [4], and material processing [5] have exploited this new light source for many yearsand even nowadays fields of application such as home media [6] still trigger the need for low-cost and mass-producible lasers. Despite the huge demand, even today the optical spectrum between the IR and the UV is not fully accessible by laser sources. The dark lines of the rainbow spanned by the different laser types, e.g. in the regime of green light [7], can be filled by nonlinear optics. Optical parametric os- cillation [8] and second harmonic generation [9] are two possible mech- anisms to convert light to different wavelengths that both benefit from quasi phase matching [10, 11]. The two mechanisms are based on high- quality non-linear crystals. One of the most important non-linear-optical materials is lithium nio- bate [12, 13], due to its ease of fabrication, robustness, transparency in the visible-to-infrared and excellent nonlinear properties [14]. Lithium niobate possesses a shift between the distribution of the cations and the anions, at room temperature, along the crystalographicc-axis, leading to a so-called spontaneous polarization. Although, it was once considered to be a frozen ferroelectric [15], during the last couple of decades the possibility to influence the orientation of the spontaneous polarization of lithium niobate and therefore its ferroelectricity has become a vivid field of research [16, 17]. The possibility to tailor ferroelectric domains allows for a wealth of applications [18, 19] and therefore it is of special interest. But many of the applications of lithium niobate involve high light intensi-

1

I RTNTIONODUC

ties. Since the material is photorefractive, at high light intensities the beam profile is distorted, which is then called optical damage [20]. Doping of lithium niobate with magnesium suppresses the undesirable optical dam- age [21], making it usable for high-power applications, but it also leads to some new challenges in terms of domain structuring. Several techniques have been developed to generate ferroelectric do- main structures in lithium niobate crystals [2224]. Currently, the most common method for ferroelectric domain patterning in bulk crystals is ap- plying an electric field by structured electrodes located on one of thec- faces of the crystal [25]. The locally modulated field causes local reorient- ing of the spontaneous polarization. Unfortunately, this method becomes more challenging for magnesium-doped material [26] and the smallest bulk domain structures that would be desirable [18] have not been real- ized yet. In this thesis the issue of tailoring ferroelectric domain structures is ap- proached from two sides: interaction of defect structures inside the crystal with growing ferroelectric domains is investigated and also actual domain patterning on all crystal faces by different methods is performed. Special emphasis is given to the Mg-doped material. The fundamental under- standing and the methods of domain patterning developed in this thesis are then used to obtain tailored domain structures that meet the require- ments of their intended application in non-linear optics.

2

Chapter

2

Fundamentals

The versatile, tunable features of lithium niobate enable various optical applications. In this chapter the general properties of lithium niobate are reviewed first and then the ferroelectric properties of this material, which lie within the scope of this thesis, are summarized.

2.1

2.1.1

Lithium niobate crystals  general properties

Crystal structure and symmetry

Lithium niobate (LiNbO 3 ) is a birefringent, piezo- and pyroelectric crys- tal that exists in its ferroelectric phase below the Curie temperature of T C = crystal structure, which belongs to the point group1411 K The [13]. 3m[12], is invariant under a 120°rotation and exhibits a mirror plane con- taining the rotation axis, thec-axis of the crystal. In this work thez-axis of the cartesian coordinate system is parallel to thec-axis. The orientation of thec-axis is given by the position of the two cations lithium (Li) and nio- bium (Nb) as well as the vacancies () with respect to the oxygen planes, where the+cdirection is defined by the sequence: ..., Li, Nb,, Li, Nb,, ... (Fig. 2.1). The displacement of the cations relative to the center between two oxygen planes along the polarc-axis gives rise to the dipoles leading to the spontaneous polarization [27]:

P S =0.71 C/m 2 . (2.1) When LiNbO 3 crystals are heated up toT C , which lies below the melt- ing temperature of 1520 K [13], the displacement and therefore the spon- taneous polarization vanishes. A second-order phase transition from fer- roelectric to paraelectric takes place [28].

3