Magneto-optical transmission of colloidal molybdenum disulphide

Home

Search

Collections

Journals

About

Contact us

My IOPscience

Magneto-optical transmission of colloidal molybdenum disulphide

This article has been downloaded from IOPscience. Please scroll down to see the full text article. 1974 J. Phys. D: Appl. Phys. 7 2483 (http://iopscience.iop.org/0022-3727/7/18/307) View the table of contents for this issue, or go to the journal homepage for more

Download details: IP Address: 14.139.242.99 The article was downloaded on 20/06/2013 at 11:26

Please note that terms and conditions apply.

J. Phys. D: Appl. Phys., Vol. 7, 1974. Printed in Great Britain.

01974

Magneto-optical transmission of colloidal molybdenum disulphide R V Mehta and H S Shah Department of Physics, S V Regional College of Engineering and Technology, Surat, India Received 11 February 1974, in final form 4 June 1974 Abstract. The optical transmission through oil-dispersed colloidal molybdenum disulphide has been investigated in a magnetic field at two different wavelengths. The experimental results show that the optical transmission strongly depends on the state of polarization of incident light. The results are interpreted in terms of the contribution due to magnetic dipole radiation in addition to that due to electric dipole.

1. Introduction The optical transmission of certain colloids and suspensions changes when placed in a magnetic field (Heller 1939, Stott 1949, Naik and Desai 1965). A study of transmission changes as a function of applied field, wavelength and state of polarization of incident light provides useful information about the size, size distribution, magnetic and optical properties of the dispersed particles (Heller 1939, Shah et al 1968, Mehta 1973). The method is useful when inference regarding the dispersed phase is to be obtained in situ and when other direct methods are not readily accessible. In our previous papers we have reported our investigations on several ferrites (Desai et al 1969, Dave et al 1968), colloidal graphite (Shah et al 1968) and suspensions of bentonite (Mehta et al 1971b, Mehta 1973). This paper describes investigations of colloidal molybdenum disulphide in oil. The colloid was obtained from Acheson Colloids Co. The optical and magnetic properties of large crystals of MoSz were studied earlier (Tyndall 1923, Dutta 1945, Paul 1967) but those of small particles have not been found in the literature. An MoSz crystal has unusually large optical constants, so it is interesting to study the magnetooptical transmission of this colloid. 2. Experimental details The experimental assembly is shown in figure 1. Light from monochromator of the UNICAM-SPSOO spectrophotometer passes through a quartz collimating lens L1, a Glan-Thomson prism P, a specimen cell C and a receiving lens Lz. The Glan-Thomson prism was cemented with linseed oil in order to extend the investigation to the violet end of the spectrum. The specimen cell was placed between two pole pieces of an electromagnet. Three right-angled prisms facilitated observations parallel to the direction of the field. It has been shown earlier that transmission changes in the longitudinal direction

248 3

2484

R V Mehta and H S Shah

I

Figure 1. Experimental set-up to study the magneto-optical transmission changes: S, slit of Unicam Sp 500 Spectrophotometzr; L1, collimating lens; I, iris diaphragm; P, Glan-Thomson prism; C, specimen cell; L2, receiving lens; PM, photomultiplier; A, difference amplifier; G, multiflex galvanometer; EM, electromagnet; RI, Rz, R3, right-angled prisms.

are independent of the state of polarization of the light (Mehta et ai 1971b, Foweraker and Jennings 1973). Hence unpolarized light was used for these measurements. Transmitted light was detected by a photometer which consisted of a photomultiplier type M 12 FQS 35 (Zeiss), a difference amplifier with a voltage gain of 140 (figure 2) and a multiflex galvanometer with a maximum sensitivity df 10-9 A per division. The multiplier was kept at a suitable distance from the electromagnet, so that the response of the multiplier was not affected by the field. The resistor chain of the multiplier was directly connected to the base pin, and resistances were impregnated with wax to mini* " mize leakage due to humidity. -This unit was found to be capable of detecting a photo-

Siam

-

+36V 1

Magneto-optical transmission of colloidal molybdenum disulphide

2485

current of the order of 10-11 A. The linearity of the response of the multiplier was checked for the ranges of intensity used in this study. The electric vector of the incident light was oriented in the desired direction with respect to the direction of the applied field using a technique based on magnetooptical effects (Mehta et al 1971a). A cell containing dispersion of Aquadag graphite (Acheson Colloids CO)was placed between the pole pieces of the electromagnet. A monochromatic light beam was incident on a polarizer. After passing through the dispersion, the transmitted beam was viewed through an analyser. The polarizer and the analyser were initially crossed. When the field was applied the field of view brightened because of the induced dichroism. The orientation of the polarizer was found for which the field of view remained dark with or without field. This would be the case when electric vector was either parallel or perpendicular to the applied field. Since the Aquadag graphite showed fairly large dichroism at the moderate field, it was found to be more suitable for this purpose. The changes in the transmission through the sample under investigation was measured as follows. Initially, with the slit of monochromator closed, the current in the galvanometer was adjusted to zero with the help of difference amplifier. The slit was then gradually opened, and the current in the galvanometer was balanced to zero. The total output current of the amplifier corresponding to a deflection d in the galvanometer was noted. The slit width was kept at a minimum, and most of the investigations were carried out with a slit width of 0-1 mm. Band widths corresponding to this slit width at 4000 and 6500 AU were found to be 10 and 30 AU respectively. The transmittivity Z/Zo, where l o is the intensity of the incident light and Z is that of transmitted light, of the colloid under the zero field was also measured at the same slit width. The method minimized the error arising through the finite bandwidth of the light beam. When the field was applied, the change in the transmission was then indicated by a corresponding deflection Ad in the galvanometer. Since Ad could be measured accurately up to of the scale division, transmission changes up to 0.05% could be measured. This quantity is expressed as a ratio of extinction cross section by the following relation :

a

CF QF=-=l---

CO

In (1 + d / A d )

CO

where CF is the extinction cross section of the colloid under the applied field and COis the extinction cross section of the colloid under zero field. The optical transmission QF was studied for three cases: (i) with the magnetic field applied transverse to the direction of propagation of the incident light, the light being linearly polarized with its electric vector parallel to the direction of the field (QF= QL); (ii) with the magnetic field as above and the light being linearly polarized with its electric vector perpendicular to the field (QF = QR); (iii) with the magnetic field applied along the direction of propagation of incident light, the changes observed in this case being independent of the state of polarization of the incident light (QF= QK).

3. Results The colloidal MoS2 was diluted in distilled kerosene, and a diluted sample was used for the present investigation. Beer’s law was verified for the concentration used in the study.

2486

R V Mehta and H S Shah

The changes in optical transmission in the magnetic field are shown in figure 3. Compared to our previous investigations (Shah et all968, Desai et aZ1969), transmission changes observed in this dispersion are small. Moreover, the dipole criteria Q R = QK and ( Q L - ~ ) = ~ ( ~ - Q R(Shah ) et al 1968, Mehta et al 1971b) do not hold for this sample. In order to investigate the dependence of the above criteria on the particle size, we determined the average particle size in the dispersion from the extinction curve at zero field strength.

1.06

1

1.04

0.98.

0.96.

Figure 3. Optical transmission for MoSz diluted in kerosene A = 5500 au (. . .) and 5890 au (-).

The extinction curve was plotted from the spectrophorometric study (figure 4). In the region 6000-10 000 AU the extinction coefficient is approximately proportional to A-4, and MO% crystals are transparent beyond 7000 AU (Tyndall 1923). Hence it is easy to determine the particle size from extinction data: (1)

Cext=CscafCabs.

Beyond 7000 AU Cabs=()

and

I

where M(g cm-3) is the concentration, p is the density of MO&, A’( = ho)is the wavelength of the light in the medium, V is the volume of the particle, and m( =n/no) is the relative refractive index of the dispersion, n and no being the refractive indices of the dispersed phase and the medium. Since MoS2 decomposes, even at a moderate temperature, the concentration of molybdenum, was estimated by evaporating to dryness a definite volume of the colloid

Magneto-optical transmission of colloidal molybdenum disulphide

2487

0.6 0.82 \

0.4 0.74 0.2

0,66 0:6

'

'

018

.

I.o

0.58

0.50

c

2

0.36

0.40

0.44

0.48

Mpm)

Figure 4. Curve of (Cext)oagainst H for MoSz diluted in kerosene.

and weighing the residue on a microbalance. The concentration of MoS2 was then determined from stoichiometric formula and was found to be 5.5 x 10-6 g cm-3. At A = 8000 AU, Csca=0.0672 (figure 4). Substituting these values and the value of p, IZ (Tyndall 1923), no, etc in equation (2) we obtain V=lO-15cm3 or 2r=0.12 pm. In the absence of any data on polydispersity, shape etc, this value is only a fair average. It only indicates that the particles are not very large. Thus the dipole criteria for magneto-optical effects do not hold for the colloidal MOSZparticles, even though the particles are as small as 0.12 pm in diameter. This is surprising in the light of our previous study, eg graphite dispersions in which particles are about five times larger than in the present case the dipole criteria is followed (Shah et a1 1968). The refractive index of MOSSis abnormally large (Tyndall 1923). This large value of real part of refractive index may account for this anomaly. For the limiting case of large refractive index, ie for m = CO, the contribution due to magnetic dipole scattering is significant in addition to electric dipole scattering (Hulst 1957). The contribution of electric and magnetic dipole scattering in the present case can be taken into consideration as follows. Let both the electric and magnetic fields of the incident wave be oriented along the axes of an ellipsoid. Then the corresponding depolarization factors (Le)j and (Lm)j are given by (Hulst 1957)

and since the magnetic permeability is zero,

2488

R V Mehta and H S Shah

where q S ( j = ‘0’or ‘e’) are the principal components of polarizability of the particle. Subscripts ‘e’ and ‘0’respectively indicate the component along and perpendicular to symmetry axis, and superscripts ‘e’ and ‘m’ indicate the electric and magnetic fields of the incident light. Since the electric and magnetic dipoles are induced by the field of the incident light only, the two radiations are in phase and their amplitudes are to be added to obtain the amplitude matrix S(0). Proceeding in a similar way to that in which we derived the expressions on the basis of Rayleigh theory (Shah et al 1968), the following expressions are obtained for the present case: (Cext)L

= (Cext)Le

(Cext)R

= (Cext)Re

(Cext)K

= (Cext)Re

+ + +

(Cext)Rm

(5)

(Cext)Lm (Cext)Rm

where (Cext)Le is the extinction cross section due to the electric dipole when the symmetry axis of the particle is parallel to the electric vector of the incident light and (Cext)Rm is the contribution due to the magnetic dipole radiation to the total cross section when the symmetry axis of the particle is perpendicular to the electric vector of the incident light. The two other cross sections (Cext)Re and (Cext,)Lm can be defined in a similar way. The above expressions can be used to give the following relations : (QL-~)+(QR-~)+(QK-~)=O

( i e x Q F = ~ , F=L, R, K)

(6)

and Q R > QK.

Thus if the contribution due to magnetic dipole radiation is significant, the results should follow the above criteria. It can be seen from figure 3 that in the case of colloidal MoSz ~ table 1). Q R > QK and Q F = (see Table 1

5890Au 8000 7500 7000

3.001 3.013 3.003

5500 AU 7000 6000 5000

3.015 3.003 3.025

Thus the magneto-optical effect in colloidal MoSz follows both the criteria given in equation (6). 4. Conclusions

The present study shows that a contribution due to a magnetic dipole can account for the observed anomaly in the case of MoSz colloid. Crandle (1913) reported that MoSz

Magneto-optical transmission of colloidal molybdenum disulphide

2489

crystals are weakly birefringent, but since no data were available it was not possible to determine the relative contribution due to the shape of the particles and due to the intrinsic optical anisotropy of the particks. The method described earlier by the authors (Mehta et a1 1971a,b) to determine the shape factor from the magneto-optical data is also not applicable here, since the simple electric dipole theory does not seem to hold good for the MoS2 particles. It is necessary to develop some alternative method to determine the shape of the particles from magneto-optical data. Our equation (6) may prove helpful for this purpose. References Crandle 1913 Phys. Rev. 2 343 Dave M J, Mehta RV, Shah H S, Desai J N and Naik Y G 1968 Indian J. Pure Appl. Phys. 6 364 Desai J N , Naik Y G, Mehta R V and Dave M J 1969 Indian J. Pure Appl. Phys. 7 534 Dutta A K 1945 Indian J. Phys. 19 225 Foweraker A R and Jennings B R 1973 Appl. Optics 12 1983 Heller W 1939 Properties Magneto-optiques des Solutions Colloidales (Paris: Hermann) Mehta R V 1973 J. Colloid Interface Sci. 42 165 Mehta R V and Dave M J 1971 Rev. Sci. Instrum. 42 1074 Mehta RV, Dave M J and Desai J N 1971a Appl. Optics 10 2786 Mehta RV, Shah H S and Dave M J 1971b J. Colloid Interface Sci. 35,41 Naik Y G and Desai J N 1965 Indian J.,Pure Appl. Phys. 3 27 Paul D 1967 Indian J. Phys. 41 943 Shah HS, Desai J N and Naik Y G 1968 Indian J. Pure Appl. Phys. 6 282 Stott FD 1949 Proc. Phys. Soc. B 62 418 Tyndall J 1923 Phys. Rev. 21 162 Van De Hulst H C 1957 Light Scattering by Small Particles (New York: Wiley) p 75