J Mater Sci: Mater Electron (2012) 23:739–745 DOI 10.1007/s10854-011-0481-z

Nonlinear optical characterization and measurement of optical limiting threshold of CdSe quantum dots prepared by a microemulsion technique S. Mathew • Amit D. Saran • Santhi Ani Joseph • Bishwajeet Singh Bhardwaj • Deep Punj • P. Radhakrishnan V. P. N. Nampoori • C. P. G. Vallabhan • Jayesh R. Bellare

•

Received: 24 May 2011 / Accepted: 26 July 2011 / Published online: 13 August 2011 Ó Springer Science+Business Media, LLC 2011

Abstract CdSe quantum dots prepared by micro emulsion technique shows quantum confinement effect and broad emission at 532 nm. These quantum dots have about 4.35 nm size, and they exhibit good nonlinear effects which are measured using z-scan technique. The samples have a reverse saturation in the nonlinear absorption as nonlinear optical absorption coefficient b is 2.545 9 10-10 W m-1 and nonlinear optical refraction coefficient n2 is -1.77 9 10-10 esu. The third-order nonlinear optical susceptibility is found to be 4.646 9 10-11 esu and also the figure of merit is 2.01 9 10-12 esu m. The optical limiting threshold which is found to be 0.346 GW/cm2 makes it a good candidate for device fabrication.

S. Mathew (&)  S. A. Joseph  P. Radhakrishnan  V. P. N. Nampoori  C. P. G. Vallabhan International School of Photonics, Cochin University of Science and Technology, Kochi 682022, India e-mail: [email protected] S. Mathew  B. S. Bhardwaj  D. Punj  P. Radhakrishnan Centre of Excellence in Lasers and Optoelectronics Sciences, Cochin University of Science and Technology, Kochi 682022, India A. D. Saran  J. R. Bellare Department of Chemical Engineering, Indian Institute of Technology Bombay, Mumbai 400076, India J. R. Bellare School of Biosciences and Bioengineering, Indian Institute of Technology Bombay, Mumbai 400076, India

1 Introduction Semiconductor quantum dots are very important class of materials requiring detailed studies to understand the optical properties [1–3]. They exhibit strong size dependence of spectral properties [4–7] and hence are suitable for a vast range of applications, such as light-emitting devices, non-linear optical devices, solar cells, and as fluorescent bio-labels [8–12]. Linear optical properties of the CdSe quantum dots have been extensively studied [13– 15]. The studies of nonlinear optical processes in photonic materials arise from a major field of research considering the importance of their technological applications, especially in areas such as passive optical power limiting, optical switching, and in the design of logic gates [16–18]. However, only a few measurements are reported regarding nonlinear optical properties of the CdSe quantum dots through pump probe and degenerative four wave mixing, but they give no adequate conclusive and quantitative results. Therefore for systematic and quantitative analysis, third-order optical nonlinearity of the CdSe quantum dots has been investigated using z-scan method in this work [19, 20]. When the size of quantum dot is reduced to the order of an exciton Bohr radius aB, quantum size effects appear and drastic changes in optical properties are expected. The quantum confinement effect in semiconductor nanocrystals can be classified into two regimes, i.e., the strong and weak confinement regimes, with respect to the ratio of nanocrystal radius r to aB [21]. In the strong-confinement regime, the photo-excited electron and hole are individually confined. Theoretical and experimental works have revealed that the state-filling effect accounts for the nonlinearity in this regime. In the case of strong confinement of electrons and holes, i.e., when the quantum dot diameter is smaller in size than an exciton, large optical

123

740

nonlinearities are expected [22]. In the present study, the CdSe quantum dots are prepared by micro emulsion technique. The microemulsion synthesis method is simple, inexpensive, highly reproducible, and offers great control over the size and shape of nanoparticles while having high thermal stability. As a result, water-in-oil microemulsions have been used for synthesis of an extensive range of nanomaterials of metals [23], semiconductors [24], metal carbonates [25], etc. The quantum confinement effects resulting from the nanometer size of the sample would enlarge the band gap and enhance the nonlinear optical response. CdSe quantum dots within the regime of a strong quantum confinement have been investigated as a case with respect to some nonlinear optical responses including two photon absorption, thermal lens effects, free-carrier absorption etc. The work reported in this article shows that the CdSe quantum dots of size coming in the strong confinement regime, give a broad emission around 532 and 566 nm and possess a good optical nonlinearity. The nonlinear refraction coefficients and nonlinear absorption coefficients of the CdSe quantum dots have been measured at the wavelength of 532 nm. The phenomenon of optical limiting has generated great interest in the protection of eyes and the sensors from intense laser beam. Optical limiting happens when the absolute transmittance of a material decreases with an increase in input fluence. One mechanism for optical limiting is provided by reverse saturable absorption (RSA), in which the excited state absorption cross section is higher than the ground state absorption cross section. The opticallimiting effects of CdSe quantum dots have also been investigated here with nanosecond laser pulses.

2 Experimental details The details of sample preparation are briefly described below. The starting materials are the surfactant, dioctyl sulfosuccinate sodium salt (Aerosol-OT, or AOT) of 99% purity, Cadmium nitrate and Zinc Nitrate of AR grade and n-heptane (extra pure, 99%). These materials are purchased from S.D. Fine (India) chemicals. Ammonium sulfide ((NH4)2S) (25% aqueous solution), and Sodium Sulfite (anhydrous) were purchased from Alfa Aesar, (UK). Selenium powder (black, GR grade) of 99.5% purity was supplied by Kemie Labs, India. All the chemicals were used without any further purification. Ultra-pure water (Milli-Q, Millipore) was used throughout the experiments. The synthesis of the CdSe quantum dots is carried out as per a simplified microemulsion method reported recently by our group [26]. In a typical experiment, 6.32 g of selenium powder is added to 200 mL of Na2SO3 solution

123

J Mater Sci: Mater Electron (2012) 23:739–745

(0.2 M) with continuous nitrogen bubbling. The pH is maintained at about 10 by adding dilute NaOH solution. The reaction is continued at 80 °C for about 6 h. The reaction mixture is then filtered and the clear solution is kept in dark. The unreacted selenium powder is weighed and, it is found that stochiometrically about 60% of sodium sulfite is consumed. Accordingly, the concentration of the Na2SeSO3 solution should be 0.12 M. This solution is further diluted to give a stock solution of 0.1 M Na2SeSO3. This solution needs to be kept in dark and should be used up within 4–5 days. Reverse micelle system is prepared by dissolving AOT in n-heptane (0.1 M). The reverse micelles are then swollen with appropriate aqueous precursor solutions thus forming microemulsions (lE). Microemulsion of Cd(NO3)2 (0.3 M) and Na2SeSO3 (0.1 M), are prepared at water-to-surfactant molar ratio, R = 4. For example, for preparing microemulsion of Cd(NO3)2 at R = 10, we need to add 90 l of aqueous solution to 5 mL of AOT/n-heptane solution. The microemulsion is stirred with vortex mixer till it becomes completely clear. The two microemulsions (Cd(NO3)2 and Na2SeSO3) thus prepared and then vortexmixed, thereby leading to the formation of CdSe inside the reverse micelles. Absorption spectra of QDs in microemulsion were recorded using UV–Visible Spectrophotometer (Nicolet Evolution 300). The microemulsions as prepared were used directly for absorption studies. For QD microemulsions, blank microemulsion (0.1 M AOT in n-heptane stock solution) was used as a reference. Fluorescence spectra of QDs in microemulsion were obtained on a Perkin–Elmer LS55, fluorescence spectrometer. The excitation wavelength was kept at 390 nm, and the emission spectra were recorded. The Electron micrograph images of QDs in microemulsion were taken using Transmission electron microscope (Philips Technai G2 at 120 kV). For preparation of grid from microemulsion, a drop of microemulsion is first loaded on a carbon-coated grid, which is allowed to dry for about 15 min. This grid is then dipped in pure n-heptane in two stages of 30 s each so as to remove surfactant with minimum loss of nanoparticles and then in water again for 30 s to remove unreacted precursor salts. Then the grid is dried under IR lamp for about 2 h, after which it can be observed under the microscope. For nonlinear characterization, we have used the single beam z-scan technique with nanosecond laser pulses to measure nonlinear optical absorption and refraction properties of the CdSe quantum dots. Z-scan technique developed by Sheik Bahae and his co-workers is a single beam method for measuring the sign and magnitude of nonlinear refractive index, n2, and the nonlinear absorption coefficient b and the method has good sensitivity comparable to interferometric methods. The experimental set up is as

J Mater Sci: Mater Electron (2012) 23:739–745

741

Fig. 1 Experimental setup to extract z-scan data

shown in Fig. 1. A Q-switched Nd:YAG laser (Spectra Physics LAB-1760, 532 nm, 7 ns, 10 Hz) is used as the light source. The sample is moved in the direction of light propagation near the focal spot of the lens with the focal length 200 mm. The radius of the beam waist is calculated to be 43.3 lm. The Rayleigh length, z0 is estimated to be 11 mm, which is much greater than the thickness of the sample cuvette (1.1 mm), an essential prerequisite for z-scan experiments. The linear transmittance of the farfield aperture S, defined as the ratio of the pulse energy passing the aperture to the total energy, is measured to be approximately 0.08. The z-scan system is calibrated using CS2 as a standard. The transmitted beam energy, reference beam energy and their ratio are measured simultaneously by an energy ratiometer (Rj7620, Laser Probe Corp.) having two identical pyroelectric detector heads (Rjp735). The effect of fluctuations of laser power is eliminated by dividing the transmitted power by the power obtained at the reference detector; both being measured using identical photo detectors. The data are analyzed by using the procedure described by Sheik-Bahae et al. [27].

3 Results and discussions The absorption spectrum of the CdSe quantum dots is given in Fig. 2. It is clear from the spectrum that the first exciton peak is sharp which indicates a narrow size distribution. Due to the narrow size distribution, the differences in band gap energy of different sized particles will be very small and hence most of the electrons will get excited over a smaller range of wavelengths. In addition, if there is a narrow size distribution, the higher exciton peaks are also seen clearly and it is also present here in the absorption spectrum. The optical absorption bands in the spectrum are fitted to two Gaussian bands. The peaks of the two bands are at 2.21 eV (denoted as 1 in Fig. 2) which is the 1se–1sh transition and other at 2.63 eV (denoted as 2 in Fig. 2) corresponding to the same transition with the hole in the spin–orbit-split valence

Fig. 2 Absorption spectrum of the CdSe quantum dots

band [14]. Comparing the absorption peak maximum at 566 nm corresponding to the first excitonic transition with the curves reported by Peng and co-workers [28], the quantum dot diameter is calculated to be 3.41 nm. Figure 3a shows the TEM image of the CdSe quantum dots at R = 4. Two important observations can be derived from these micrographs. These CdSe QDs are roughly spherical having mean size of 4.35 nm. Figure 3b gives the electron diffraction pattern for CdSe quantum dots. The electron diffraction pattern indicates highly crystalline structures having fcc pattern. Figure 4 shows the fluorescence spectrum of the CdSe quantum dots under the excitation of 390 nm using Perkin–Elmer LS55, fluorescence spectrometer. There is a broad emission around 532 nm and it is assigned to excitonic emission from these nanoparticles corresponding to that reported for the 1se–1sh excitonic state of the CdSe quantum dots [13]. Figure 5a gives the open aperture z-scan trace of CdSe quantum dots at a typical fluence of 392 MW/cm2. The open-aperture curve exhibits a normalized transmittance valley, indicating the presence of a RSA in the quantum dots. The open-aperture z-scan scheme was used to measure nonlinear absorption coefficient, b and the imaginary part of the third-order nonlinear optical susceptibility, Im v(3). The data are analyzed by using the procedure described by Sheik-Bahae et al. [27] for a two photon absorption process; the nonlinear absorption coefficient is obtained by fitting the experimental z-scan plot to Eq. 1, Z1 C 2 T ðzÞ ¼ lnð1 þ q0 et Þdt ð1Þ q0 p 1

where q0 ðz; r; tÞ ¼ bI0 ðtÞLeff and Leff ¼ ð1  eal Þ=a) is the effective thickness with linear absorption coefficient, a and I0 the irradiance at the focus. The solid curves in Fig. 3 are

123

742

Fig. 3 a TEM picture of the CdSe quantum dots. b Electron diffraction pattern of the CdSe quantum dots

Fig. 4 Fluorescence spectrum of the CdSe quantum dots under the excitation wavelength of 390 nm

123

J Mater Sci: Mater Electron (2012) 23:739–745

the theoretical fit to the experimental data. The calculated value for b is 2.545 9 10-10 W m-1. Im v(3) is related to b through the equation Imð3Þ ¼ n20 c2 b=ð240p2 xÞ; (in esu) where n0 = 2.34 is the linear refractive index of CdSe, e0 is the permittivity of free space, c is the velocity of light in vacuum, x is the angular frequency of the radiation used. The value of imaginary part of v(3) is found to be 1.5 9 10-11 esu corresponding to input intensity of 392 MW/cm2. In addition to the measurement of nonlinear absorption, the laser induced fluorescence is also investigated. Light emission is collected from the side of 1 cm 9 1 cm cuvette containing sample using a collecting optical fiber coupled to a monochromator-CCD system (Acton Spectrapro). Figure 5b shows the laser induced fluorescence spectrum of these quantum dots excited by the 532 nm nano second laser pulses. It shows a broad emission around 568 nm which is due to the inter-band transition. The FWHM of 74 nm for this emission is primarily attributed to the size distribution of these QDs. There is a chance that radiationless relaxation effect may not be significant so that band to band transition may give significant width for the fluorescence emission process. In order to explain the optical nonlinearity of the CdSe quantum dots, the system can be considered as a two band system and the nonlinear absorption mechanism is shown in Fig. 5c. In our experimental scheme, the nonlinear absorption is mainly contributed to an excited-state absorption or free carrier absorption process, in which electrons in the valence band are excited to the conduction band energy states and as a consequence generation of free charge carriers occurs. Thus a large number of free carriers is generated by one-photon excitation by a high intensity 532 nm laser pulse. Subsequently, the free carriers formed could further absorb another photon and be excited to a higher conduction-band level, resulting in the transient absorption. Also from laser induced fluorescence Fig. 5b, it is clear that a part of the absorption leads to fluorescence. In this process, we suggest that the lower level of the conduction band may act as an intermediate state to generate the RSA. Also the size of these quantum dots (4.35 nm) comes inside the exciton Bohr diameter of 9.8 nm in the case of CdSe. i.e. in the strong confinement regime. So the exciton oscillator strength gets enhanced due to quantum confinement effect and this will also enhance the nonlinear optical property. Thus free carrier absorption as well as quantum confinement contribute to the nonlinear optical absorption of these quantum dots. The open aperture z-scan curve of the CdSe quantum dots at different input fluences are shown in Fig. 5d. We can see that nonlinear absorption is highly fluence-dependent. The values of b and Imv(3) at different fluences are shown in Table 1 This indicates that the free carrier density increases with input intensity.

J Mater Sci: Mater Electron (2012) 23:739–745

743

Fig. 5 a Open aperture z-scan plot of the CdSe quantum dots. b Laser induced flourescence under 532 nm nanosecond laser pulse. c A mechanism for nonlinear absorption in the CdSe quantum dots. d Fluence dependent of transmission of the CdSe quantum dots

Table 1 The variation of b with different input intensities for the CdSe quantum dots Intensity (MW/cm2)

b (m/W) 9 10-10

Im(v3) (esu) 9 10-12

212

0.4894

2.88

247

1.478

8.692

392

2.545

15

Figure 6 shows the closed-aperture z-scan trace of the CdSe quantum dots for input fluence of 392 MW/cm2. The closed-aperture curve displayed a peak to valley shape, indicating a negative value of the nonlinear refractive index n2 and it is the self-defocusing nonlinearity. For samples having sizable refractive and absorptive nonlinearities, closed-aperture measurements contain contributions from both the intensity-dependent changes in the transmission

and in the refractive index. By dividing the normalized closed-aperture transmittance by the corresponding normalized open-aperture data, we could retrieve the phase distortion created due to the change in refractive index; this result is depicted in Fig. 6. It is observed that the valley– peak of the closed-aperture z-scan satisfied the condition Dz = 1.7 z0, thus confirming the presence of cubic nonlinearity [27]. The value of DTp-v i.e., the difference between the peak and valley transmittance, could be obtained by the best theoretical fit from the results. The nonlinear refractive index n2 and the real part of the thirdorder nonlinear susceptibility, Re v(3) are given, respectively, by Eqs. 2 and 3. n2 ðesuÞ ¼

cn0 kDTpv 40p2 0:812ð1

 SÞ0:25 Leff I0

ð2Þ

123

744

Fig. 6 Closed aperture z-scan data of the CdSe quantum dots

n0 n2 ðesuÞ ð3Þ 3p The calculated value of n2 is -1.77 9 10-10 esu and the value of real part of v(3) is found out to be 4.397 9 10-11 esu. The third-order nonlinear susceptibility of the CdSe quantum dots is calculated from real and imaginary part of v(3) by the relation of [Re v(3) ? Im v(3)]1/2 and it is found to be 4.646 9 10-11 esu. Its figure of merit is found to be (v(3)/a) 2.01 9 10-12 esu m. It is important to note that certain representative third-order nonlinear optical materials, such as Ag2S/CdS nanocomposites [29], metallophthalocyanines [30], porphrins [31], organic dyes [32], organic polymers [33], organic coated quantum dots [34], metal clusters [35], etc., yielded values of the order of 10-10–10-14 esu for v(3) at a wavelength of 532 nm. These values are in Re vð3Þ ¼

J Mater Sci: Mater Electron (2012) 23:739–745

the right order of magnitude in comparison to the value of v(3) obtained in the present investigation. To study the optical limiting property of the sample, the nonlinear transmission of the sample is measured as a function of input fluence. Optical power limiting is effected through the nonlinear optical processes of the sample. An important term in the optical limiting study is the optical limiting threshold. Optical limiters are eventually those systems which transmit light at low input fluence or intensities, but become opaque at high inputs. The optical limiting property is mainly found to be absorptive nonlinearity, which corresponds to the imaginary part of thirdorder susceptibility [36]. From the value of fluence at focus, the fluence values at other positions could be calculated using the standard equations for Gaussian beam waist. Such plots give a better comparison of the nonlinear absorption or transmission in the sample and are generated from z-scan trace. Figure 7 illustrates the optical limiting response of the sample. The line in the figure indicates the approximate fluence at which the normalized transmission begins to deviate from linearity. The fluence value corresponding to the onset of optical limiting (optical limiting threshold) is found to be 0.346 GW/cm2. Thus these quantum dots can be used for optical power limiting at high laser fluences.

4 Conclusions CdSe quantum dots of average size 4.35 nm were prepared by microemulison method. Broad luminescence band around 532 nm is observed for these quantum dots. The nonlinear properties of these nanoparticles are found out by z-scan method. Nonlinear optical properties of these CdSe quantum dots are investigated and optical power self-limiting application is studied. The optical nonlinearity of these quantum dots is found to be arising from free-carrier absorption as well as quantum confinement. The thirdorder nonlinear optical susceptibility is found to be 4.646 9 10-11 esu at input intensity of 392 MW/cm2. The optical limiting threshold is found to be 0.346 GW/cm2. Acknowledgments MS acknowledges CELOS (funded by UGCGovernment of India) for research fellowship. The authors are grateful to CELOS for some of the experimental facilities.

References

Fig. 7 Optical limiting of CdSe the quantum dots

123

1. K. Sebald, P. Michler, T. Passow, D. Hommel, G. Bacher, A. Forchel, Appl. Phys. Lett. 81, 2920 (2002) 2. C. Tansakul, E. Lilie, E.D. Walter, F. Rivera III, A. Wolcott, J.Z. Zhang, G.L. Millhauser, R. Braslau, J. Phys. Chem. C 114, 7793–7805 (2010) 3. D.R. Baker, P.V. Kamat, Langmuir 26(13), 11272–11276 (2010)

J Mater Sci: Mater Electron (2012) 23:739–745 4. Li. L-S, J. Hu, W. Yang, A. Paul Alivisatos, Nano Lett. 1(7), 349–351 (2001) 5. F. Wu, J.W. Lewis, D.S. Kliger, J.Z. Zhang, J. Chem. Phys. 118(1), 12–16 (2003) 6. S. Arora, S. Sundar Manoharan, J. Physics Chem. Solids. 68, 1897–1901 (2007) 7. N. Pradhan, D. Goorskey, J. Thessing, X. Peng, J. Am. Chem. Soc. 127, 17586–17587 (2005) 8. V.L. Colvin, M.C. Schlamp, A.P. Allvisatos, Nature 370, 354–356 (1994) 9. G.C. Bhar, G.C. Ghosh, IEEE J. Quantum Electron. 16(8), 843–846 (1980) 10. Y.-L. Lee, Y.-S. Lo, Adv. Funct. Mater. 19, 604–609 (2009) 11. Y. Zhou, F.S. Riehle, Y. Yuan, H.-F. Schleiermacher, M. Niggemann, G.A. Urban, M. Kru¨ger, Appl. Phys. Lett. 96, 013304 (2010) 12. C.-P. Huang, H.-W. Liu, C.-Y. Tsao, L.-T. Yin, S.-F. Chiu, T.-M. Chen, Sens. Actuators B: Chem. 108(1–2), 713–720 (2005) 13. M.G. Bawendi, W.L. Wilson, L. Rothberg, P.J. Carroll, T.M. Jedju, M.L. Steigerwald, L.E. Brus, Phys. Rev. Lett. 65, 1623–1626 (1990) 14. B.O. Dabbousi, M.G. Bawendi, O. Onitsuka, M.F. Rubner, Appl. Phys. Lett. 66(11), 1316–1318 (1995) 15. M.G. Bawendi, P.J. Carroll, W.L. Wilson, L.E. Brus, J. Chern. Phys. 96(2), 946–954 (1992) 16. J.M. Luther, P.K. Jain, T. Ewers, A.P. Alivisatos, Nature Materials 10, 361–366 (2011) 17. M.C. Hoffmann, B.S. Monozon, D. Livshits, E.U. Rafailov, D. Turchinovich, Appl. Phys. Lett. 97, 231108 (2010) 18. R. del Coso, J. Requejo-Isidro, J. Solis, J. Gonzalo, C.N. Afonso, J. Appl. Phys. 95, 2755–2761 (2004) 19. K.-C. Je, I.-C. Shin, J. Kim, K. Kyhm, Appl. Phys. Lett. 97, 103110 (2010)

745 20. S.H. Park, M.P. Casey, J. Falk, J. Appl. Phys. 73, 8041 (1993) 21. Y. Kayanuma, Phys. Rev. B 38, 9797 (1988) 22. P. Roussignol, D. Ricard, C. Flytzanis, Appl. Phys. B: Lasers Opt. 51, 437 (1990) 23. W. Zhang, X. Qiao, Mater. Chem. Phys. 109, 411–416 (2008) 24. M. Chu, Y. Sun, S. Xu, J. Nanopart. Res. 10, 613–624 (2008) 25. C.Y. Tai, C. Chen, Chem. Eng. Sci. 63, 3632–3642 (2008) 26. A.D. Sarana, J.R. Bellare, Coll. Surf. A: Physicochem. Eng. Aspects 369, 165–175 (2010) 27. M. Sheik-Bahae, A.A. Said, E.W. Van Stryland, IEEE J. Quantum Electron. 26, 760 (1990) 28. W.W. Yu, L. Qu, W. Guo, X. Peng, Chem. Mater. 15, 2854–2860 (2003) 29. M.Y. Han, W. Huang, C.H. Chew, L.M. Gan, X.J. Zhang, W. Ji, J. Phys. Chem. B 102, 1884 (1998) 30. J. W. Perry, L. R. Khundkar, D. R. Coulter, T. H. Wei, E. W. Van Stryland, D. J. Hagan, in Proceedings of the NATO Workshop on Organic Materials for Nonlinear Optics and Photonics, La Rochelle, 26–31 August 1990 31. G. Wood, M. Miller, A. Mott, Opt. Lett. 20, 973 (1995) 32. S.N.R. Swatton, K. Welford, S. Till, J. Sambles, Appl. Phys. Lett. 66, 1868 (1995) 33. C.S. Winter, R.J. Manning, S.N. Oliver, C.A.S. Hill, Opt. Commun. 90, 139 (1992) 34. H.P. Li, B. Liu, C.H. Kam, Y.L. Lam, W.X. Que, L.M. Gan, C.H. Chew, G.Q. Xu, Opt. Mater. 14, 321 (2000) 35. S. Shi, W. Ji, S.H. Tang, J. Am. Chem. Soc. 116, 3615 (1994) 36. F.M. Qureshi, S.J. Martin, X. Long, D.D.C. Bradley, F.Z. Heneri, W.J. Balu, E.C. Smith, C.H. Wang, A.K. Kar, H.L. Anderson, Chem. Phys. 231, 87 (1998)

123