Wave coupling theory of Quasi-Phase-Matched ... - OSA Publishing

1 downloads 0 Views 162KB Size Report
sensitive to the temperature and incident light wavelength, but it has a large ... W. L. She and W. K. Lee, “Wave coupling theory of linear electrooptic effect,” Opt.
Wave coupling theory of Quasi-Phase-Matched linear electro-optic effect Guoliang Zheng, Hongcheng Wang, and Weilong She* State Key Laboratory of Optoelectronic Materials and Technologies, Sun Yat-sen University, Guangzhou 510275, China [email protected]

Abstract: Starting from Maxwell’s equations and taking the nonlinearity of linear electro-optic effect as a perturbation, we derive general wave coupling equations of quasi-phase-matched (QPM) linear electro-optic effect. And then we use the equations to study the electro-optic effect in PPLN. The numerical results indicate that the QPM condition plays an important role in electro-optic coupling. In addition, the coupling is very sensitive to the temperature and incident light wavelength, but it has a large tolerance to the direction of incident light. ©2006 Optical Society of America OCIS codes: (190.0190) nonlinear optics; (160.2100) Electro-optical materials.

References and links 1. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18.

J. A. Armstrong, N. Bloembergen, J. Ducuing, and P. S. Pershan, “Interactions between light waves in a nonlinear dielectric,” Phys. Rev. 127, 1918-1939 (1962). K. Mizuuchi and K. Yamamoto, “Highly efficient quasi-phase-matched second-harmonic generation using a first-order periodically domain-inverted LiTaO3 waveguide,” Appl. Phys. Lett. 60, 1283-1285 (1992). S. N. Zhu, Y. Y. Zhu, and N. B. Ming, “Quasi-phase-matched third-harmonic generation in a quasi-periodic optical superlattice,” Science 278, 843–846 (1997). D. Feng, N. B. Ming, J. F. Hong, Y. S. Yang, J. S. Zhu, Z. Yang, and Y. N. Wang, “Enhancement of secondharmonic generation in LiNbO3 crystals with periodic laminar ferroelectric domains,” Appl. Phys. Lett. 37, 607–609 (1980). S. N. Zhu, Y. Y. Zhu, Y. Q. Qin, H. F. Wang, C. Z. Ge, and N. B. Ming, “Experimental realization of second harmonic generation in a Fibonacci optical superlattice of LiTaO3,” Phys. Rev. Lett. 78, 2752–2755 (1997). G. Z. Luo, S. N. Zhu, J. L. He, Y. Y. Zhu, H. T. Wang, Z. W. Liu, C. Zhang, and N. B. Ming, “Simultaneously efficient blue and red light generations in a periodically poled LiTaO3,” Appl. Phys. Lett. 78, 3006–3008 (2001). K. K. Fradkin, A. Arie, P. Urenski, and G. Rosenman, “Multiple nonlinear optical interactions with arbitrary wave vector differences,” Phys. Rev. Lett. 88, 023903 (2002). P. Xu, S. H. Ji, S. N. Zhu, X. Q. Yu, J. Sun, H. T. Wang, J. L. He, Y. Y. Zhu, and N. B. Ming, “Conical second harmonic generation in a two-dimensional χ(2) photonic crystal: a hexagonally poled LiTaO3 crystal,” Phys. Rev. Lett. 93, 133904 (2004). A. Chowdhury, H. M. Ng, M. Bhardwaj, and N.G. Weimann, “Second-harmonic generation in periodically poled GaN,” Appl. Phys. Lett. 83, 1077-1079 (2003). Y.Q. Lu, Z. L. Wan, Q. Wang, Y. X. Xi, and N. B. Ming, “Electro-optic effect of periodically poled optical superlattice LiNbO3 and its applications,” Appl. Phys. Lett.77, 3719-3721 (2000). Y. Q. Lu, M. Xiao, and G. J. Salamo, “Wide-bandwidth high frequency electro-optic modulator based on periodically poled LiNbO3,” Appl. Phys. Lett. 78, 1035–1037 (2001). K. T. Gahagan, D. A. Scrymgeour, J. L. Casson, V. Gopalan, and J. M. Robinson, “Integrated high-power electro-optic lens and large-angle deflector,” Appl. Opt. 40, 5638–5642 (2001). D. A. Scrymgeour, A. Sharan, V. Gopalan, K. T. Gahagan, and J. L. Casson, “Cascaded electro-optic scanning of laser light over large angles using domain microengineered ferroelectrics,” Appl. Phys. Lett. 81, 3140–3142 (2002). A.Yariv, “Coupled-mode theory for guided-wave optics,” IEEE J. Quantum Electron. QE 9, 919-933 (1973). W. L. She and W. K. Lee, “Wave coupling theory of linear electrooptic effect,” Opt. Commun. 195, 303-311 (2001). D. D. Wu, H. B. Chen, W. L. She, and W. K. Lee, "Wave coupling theory of the linear electro-optic effect in a linear absorbent medium," J. Opt. Soc. Am. B 22, 2366-2371 (2005). J. F. Nye, Physical properties of crystals (Oxford University Press, Oxford, 1985), Chap. 13. M. V. Hobden and J. Warner, “The temperature dependence of the refractive indices of pure lithium niobate,” Phys. Lett. 22, 243-244 (1966).

#69112 - $15.00 USD

(C) 2006 OSA

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5535

1. Introduction The “quasi-phase matching” proposed by Bloembergen et al. in 1962 provides an efficient method to match a nonlinear process in phase [1]. In order to achieve QPM, several useful and reliable techniques have been developed for modulating the nonlinear coefficient, including periodic or quasi-periodic electric-field poling of ferroelectric materials [2-8], and orientation patterning of semiconductors such as GaN [9]. The QPM technology has been widely used in nonlinear frequency conversions and receives more and more research interests [2-9]. Recently, the concept and technique of QPM were used for the linear electro-optic effect in the materials whose electro-optic coefficients are periodically modulated, and some interesting phenomena have been observed [10-13]. Lu et al. [10] used the coupled-mode theory developed by Yariv [14] to investigate the electro-optic coupling in a PPLN crystal. However, it is applicable only for some special cases (see the discussion below). On the other hand, a wave coupling theory, different from the refractive index ellipsoid and coupled-mode ones, of linear electro-optic effect was developed by She and Lee [15] and it has been generalized to the case of absorption medium [16]. However, these theories can only deal with the phenomenon of linear electro-optic effect in a single-domain bulk medium and no QPM analyses have been involved in them. In this paper, we generalize further the wave coupling theory of linear electro-optic effect to the case of QPM, then seek for the general solution of the resultant equations, and finally give some numerical analyses to the non-absorbing PPLN crystal, whose refractive indices and electro-optic coefficients are real. 2. Theory Linear electro-optic effect can be regarded as a nonlinear interaction between a light wave and an external electric field [15]. The total electric field E participating in the process of linear electro-optic effect can be expressed as E(t ) = E(0) + [Eω exp(−iωt ) / 2 + c.c.],

(1)

where [Eω exp(-iωt)/2+c.c.] is the light field with frequency ω, E(0) is the dc electric filed or slow varying electric field, and c.c. denotes the complex conjugate. In general, there exist two independent plane electromagnetic wave components for a monochromatic light wave with frequency ω propagating in a birefringent crystal, i.e., Eω = E1ω + Eω2 = E1 (r ) exp(ik1r ) + E 2 (r ) exp(ik 2 r )

(2)

where E1ω and Eω2 denote two perpendicular components when k1=k2, or two independent components experiencing different refractive indices when k1≠k2. Let E1 (r ) = ω / n1 A1 (r )a, E 2 (r ) = ω / n 2 A2 (r )b, E(0) = E 0 c ,

(3)

where a, b and c are three unit vectors and a·b=0, A1(r) and A2(r) are the normalized amplitudes of the two waves, n1 and n2 are the unperturbed refractive indices of the two waves with different polarizations. We assume that the crystal is in the clamped condition, i.e., the converse piezoelectric and photo-elastic effects are suppressed [17], and other second-order nonlinear effects are so weak because of phase mismatch that only linear electro-optic effect has to be considered. Similarly to Ref. [15], starting from Maxwell’s equations and taking the linear electro-optic effect as a perturbation, we have the following wave coupling equations under the slow varying amplitude approximation: dA1 (r ) = −iκ ⋅ g (r ) A2 (r ) exp(iΔk ' r ) − iv1 ⋅ g (r ) A1 (r ) dr dA2 (r ) = −iκ ⋅ g (r ) A1 ( r ) exp(−iΔk ' r ) − iv 2 ⋅ g ( r ) A2 ( r ) . dr

#69112 - $15.00 USD

(C) 2006 OSA

(4a) (4b)

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5536

Here, g(r) is the structure function of the material, ∆k’= k2−k2, and

κ=

k0 2 n1n2

reff 1 E0 , v1 =

k0 k reff 2 E0 , v2 = 0 reff 3 E0 , 2n1 2n2

(5)

where reffi (i=1,2,3) is the effective electro-optic coefficient [15]. Note that the wave coupling Eqs. (4a) and (4b) are different from those of coupled-mode theory developed by Yariv [14]. In Yariv’s equations, only the first term on the right-hand side of either of Eqs. (4a) and (4b) is included. In fact, the second term will influence the phase of light and cannot be ignored. If g(r) is a periodic function of r with period Λ due to periodically modulated electrooptic coefficient, it can be written as a Fourier series, g (r ) =

∑G +∞

m = −∞

m

exp(iα m r ) ,

(6)

where the mth-harmonic grating wave vector α m = 2πm / Λ is very close to −∆k’. Substituting Eq. (6) into Eqs. (4a) and (4b), and neglecting those components that make little contributions to electro-optic effect because of phase mismatch, we have dA1 (r ) ≈ −iκ q A2 (r ) exp(iΔkr ) − iv1q A1 (r ) dr

(7a)

dA2 ( r ) ≈ −iκ q* A1 (r ) exp( −iΔkr ) − iv 2 q A2 ( r ) , dr

(7b)

where Δk = Δk '+α m , and

κ q = κGm , κ q* = κG−m , v1q = v1G0 , v 2 q = v2 G0 . (8) Equations (7a) and (7b) are the resultant coupling equations of quasi-phase-matched linear electro-optic effect. Let the initial normalized amplitudes of the two waves be A1(0) and A2(0), respectively, and then the solution of Eqs. (7a) and (7b) can be obtained as follows: A1 (r ) = ρ1 (r ) exp[ iβ r + iφ1 ( r )],

(9)

A2 (r ) = ρ 2 ( r ) exp[ i ( β − Δk ) r + iφ 2 (r )],

(10)

where γ A (0) − κ q A2 (0) 2 2 ρ1 (r ) = { A12 (0) cos 2 ( μr ) + [ 1 ] sin (μr )}1 / 2 , μ

φ1 ( r ) = arg[ A1 (0 ) cos( μr ) + i

ρ 2 (r ) = { A22 (0) cos 2 ( μr ) + [

γA2 (0) + κ q* A1 (0) 2 2 ] sin ( μr )}1 / 2 , μ

φ 2 ( r ) = arg[ A2 ( 0) cos( μr ) + i μ= 1 2

(C) 2006 OSA

− γA2 (0 ) − κ q* A1 (0 )

μ

1 ( Δk + v1q − v 2 q ) 2 + 4κ q ⋅ κ q* , 2

γ = (v2 q − v1q − Δk ),

#69112 - $15.00 USD

γA1 (0 ) − κ q A2 (0 ) sin( μr )], μ

1 2

β = (Δk − v1q − v 2 q ).

sin( μ r )],

(11)

(12)

(13)

(14)

(15) (16)

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5537

Equations (9) to (16) can be used to describe the linear electro-optic effect in QPM materials with an external electric field applied along arbitrary direction. Some numerical analyses for the PPLN will be shown in section 3. 3. Numerical Analyses Without loss of generality, we take a 2.5cm long PPLN, whose poling direction is Z-axis of the crystal, as the working material. The structure function g(r) of PPLN is ⎧+ 1 g (r ) = ⎨ ⎩− 1

if

r is in

the positive domains

(17)

if r is in the negative domains, whose Fourier transform coefficient is found to be

⎧ 1 [1 − cos(2πmD ) + i sin( 2πmD )] ⎪ G m = ⎨ iπm ⎪ ⎩2 D − 1

( m ≠ 0)

(18)

(m = 0),

where D ≡ l / Λ and l is the length of one positive section. We find that the optimum value of D is 0.5 for m=1, and 0.25 or 0.75 for m=2, since |Gm| is maximized at these values of D. For the case of G0=0 (it can only be achieved with D=0.5), there is only one term left on the right hand of the coupling Eqs. (7a) and (7b), which are the same as those in the coupled-mode theory [10, 14]. Specially, if all the reciprocal vectors cannot compensate the natural dA1 (r ) dA2 (r ) ≈ ≈ 0 , meaning that the light can hardly be birefringence and v1q= v2q=0, dr dr modulated by external field. Let the angle between wave vector k1 (or k2) and Z-axis be θ. The projection of k1 (or k2) in the XY-plane is then making an angle φ with the X-axis. In the following studies, the external field direction is fixed to lie in the YZ-plane and make an angle 45° with Z-axis, i.e., c = (0, 2 / 2, 2 / 2) . The Sellmeier equations for pure LiNbO3 are given by [18]: no = 4.9130 + 2

1.173 × 10 5 + 1.65 × 10 −2 T 2 − 2.78 × 10 −8 λ2 , λ2 − (2.12 × 10 2 + 2.7 × 10 −5 T 2 ) 2

ne = 4.5567 + 2.605 × 10 − 7 T 2 + 2

0.970 × 10 5 + 2.70 × 10 −2 T 2 − 2.24 ×10 −8 λ2 , λ2 − (2.01× 10 2 + 5.4 × 10 −5 T 2 ) 2

(19)

where T is the absolute temperature and λ is the wavelength of incident light in nm. First we consider a simple structure that D is 0.5 and its first-order reciprocal vector satisfies the QPM condition when the light with wavelength λ=632.8nm propagates along the X-axis (i.e., θ=0.5π, φ=0) at temperature T=298K. Assuming the incident light is an extraordinary wave, the initial condition is given by A1(0)=0, A2(0)=1. The output intensities and phases of these two waves are shown in Figs. 1(a) and 1(b) when the QPM condition is satisfied. From the Fig. 1(a), one can find that the intensity of the extraordinary wave is totally converted to that of the ordinary wave at 0.9kV/cm. And the phases of the output waves are very regular due to the absence of v1q and v2q. Moreover, the output intensity of extraordinary (or ordinary) wave is modulated by the external electric field, which is applicable in optical modulation, optical switching, optical splitting and electric field detection, etc. If the light doesn’t propagate along the X-axis exactly, QPM condition doesn’t hold any more, because the effective period length becomes Λ ' = Λ /(sin θ cos ϕ ) and the refractive index of extraordinary wave has changed. The numerical results, for θ=0.495π and φ=0.001π, are shown in Figs. 1(c) and 1(d). One can see that the output intensities and the phases of the two waves become somewhat complicated. As a comparison, when a light propagates along the Xaxis of a single-domain bulk LiNbO3 whose structure function g(r)=1, the coupling between #69112 - $15.00 USD

(C) 2006 OSA

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5538

the two waves is very weak because of phase mismatch and the two waves behave experiencing different refractive indices independently. As mentioned above, the polar angle θ and azimuth angle φ have their influence on the QPM condition. In other words, we can tune θ and φ (by altering incident direction of light) to meet QPM condition for the case of small deviation from QPM condition. Fixing the external electric field E0 at 0.9kV/cm and varying θ from 0.49π to 0.51π and φ from -0.01π to 0.01π, we get the function relation between output intensity of ordinary wave and the angles (θ and φ), shown in Fig. 2. We can see that there exists a platform near the X-axis, which means that the coupling has a large tolerance to the change of angles θ and φ near the X-axis.

Fig. 1. Plots of intensity and phase change of the extraordinary (solid curves) and ordinary waves (dashed curves) when (a, b) the QPM condition is satisfied or not (c, d).

Fig. 2. The effect of the angles (θ and φ) on the output intensity of ordinary wave.

Besides θ and φ, temperature and the wavelength of light can also influence the QPM condition. The temperature influences the QPM condition primarily by changing the refractive indices [see Eq. (19)]. Still assuming an extraordinary light propagates along X-axis under a 0.9kV/cm applied field, we investigate the effect of the temperature T and the wavelength λ on the output intensities of the extraordinary and ordinary waves. The results are shown in Fig. 3. One can see that a 0.3K variation in temperature can lead to 50% decrease in the conversion efficiency. Therefore, a constant temperature system is necessary to achieve the stable coupling. The electro-optic coupling is also very sensitive to the incident light wavelength. The conversion efficiency drops to about 50% when the wavelength shifts by 0.1nm. This property makes it suitable to be used as a narrowband filter, whose passing frequency is governed by the QPM condition. If Λ is doubled and D=0.75, the second-order reciprocal vector will satisfy the QPM condition. In this case, v1q and v2q, which don’t appear in coupled-mode theory [10, 14], but now should be considered since G0 ≠ 0. For comparison, we study both the cases with and #69112 - $15.00 USD

(C) 2006 OSA

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5539

without consideration of v1q and v2q. The energy exchange between the extraordinary wave and the ordinary one and the phases of the two waves are shown in Fig. 4. Due to the presence of v1q and v2q, one cannot get a 100% conversion from the extraordinary wave to the ordinary one, and the phases of the two waves are irregular, which is very different from the case neglecting v1q and v2q [Figs. 4(c) and 4(d)]. So, one can draw a conclusion that v1q and v2q also play their roles in electro-optic effect.

Fig. 3. The effect of (a) the temperature T and (b) the wavelength λ on the output intensities of the extraordinary (solid curves) and ordinary waves (dashed curves).

Fig. 4. Plots of intensity and phase change of the extraordinary (solid curves) and ordinary waves (dashed curves) when (a, b) v1q and v2q are considered; and (c, d) v1q and v2q are neglected.

4. Conclusion In conclusion, we have presented a wave coupling theory of quasi-phase-matched linear electro-optic effect and found the analytical solutions for the resultant equations. The numerical investigations indicate that besides κq, κq* [see Eqs. (7a) and (7b)], v1q and v2q, the QPM condition plays an important role in the electro-optic coupling process. And the QPM condition is very sensitive to the temperature and light wavelength. Acknowledgments The authors acknowledge the financial support from the National Natural Science Foundation of China (Grant Nos. 10374121 and 10574167).

#69112 - $15.00 USD

(C) 2006 OSA

Received 21 March 2006; revised 31 May 2006; accepted 1 June 2006

12 June 2006 / Vol. 14, No. 12 / OPTICS EXPRESS 5540