w—wr AN INSTRUMENT FOR THE MEASUREMENT OF THE DEPOLARIZATION 0F RAYLEIGH SCATTERED UGHT Thesis for the Degree of M S. MECHTGAN STATE UNIVERSITY LARRY R. DOSSER 1970 IIIIIIIIIIIIIIIIIIIIIIIIIIIII WWII"llllHllHiifllUHllHIIHIIHNIIHHIUHNII 3 1293 00084 8576 LIB R A R Y Michigan Sum University 5in m: by ‘ NONE & SUNS' .- BWK BINDERY INCL L".§ffl!.‘!§9§9§3 ABSTRACT AN INSTRUMENT FOR THE MEASUREMENT OF THE DEPOLARIZATION OF RAYLEIGH SCATTERED LIGHT BY Larry R. Dosser Modifications have been made on an instrument previously built for measuring the depolarization ratio of liquids as a function of temperature. The instrument and the eXperimental techniques involved in making depolarization measurements with it are discussed in detail. The temperature dependence of the depolarization of carbon tetrachloride is extensively studied. Some data on benzene is also included. AN INSTRUMENT FOR THE MEASUREMENT OF THE DEPOLARIZATION OF RAYLEIGH SCATTERED LIGHT BY Larry RI Dosser A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF SCIENCE Department of Chemistry 1970 . U.“ \ (p '2: 2) l .4. @300 f) T ACKNOWLEDGEMENTS The author wishes to acknowledge the following persons, without whom this study would have been impossible: Mr. Russell Geyer and Mr. Charles L. Hacker, machinists; Mr. Keki P. Mistry, Mr. Andrew E. Seer, Jr., and Mr. Jerry E. DeGroot, glassblowers; Mr. Leroy B. Moy for his aid in design and construction of the electronic equipment; Mr. Stuart J. Gaumer, whose contributions to this work in the form of technical advice and assistance are too numerous to mention; and Dr. J. B. Kinsinger, for his patience and guidance. ii TABLE OF CONTENTS CHAPTER LIST OF TABLES . . . . . . . . LIST OF FIGURES. . . . . . . . I. INTRODUCTION . . . . . . . . . Purpose of this Research. . II. THE LIGHT SCATTERING APPARATUS Introduction. . . . . . . . Light Source. . . . . . . . Laser 0 O O O O I O C O 0 Alignment of the Laser Beam. Photometer. . . . . . . . . Introduction . . . . . . Temperature Control Unit Circular Polarizer . . . Polaroid Alignment . . . Polaroid Transmission. . (OCOCOU'INNN N |-“ I—‘* HHPP Lomxlfl Filters Calibration of Neutral Density Detection. . . . . . . . [\‘J }_\ N tP III. MEASUREMENT OF THE DEPOLARIZATION RATIO. Sample Cell . . . . . . . . Sample Preparation. . . . . Sample Cell Alignment . . . Measurement of PV . . . . . IV. DATA . . . . . . . . . . . . . Thermocouple Calibration. . Carbon Tetrachloride Data . Benzene Data. . . . . . . . iii NNNN LONUTIP (NU! NM #304 MN TABLE OF CONTENTS--Continued CHAPTER Page V. CONCLUSIONS. . . . . . . . . . . . . . . . . . 48 LIST OF REFERENCES . . . . . . . . . . . . . . . . . . 56 APPENDIx-—EQUIPMENT LIST 0 o o o o o o o o o o o o o o 57 iv LIST OF TABLES Page Variac Settings for Auxiliary Heater. . . . . 15 Transmission of Polaroids . . . . . . . . . . 19 Filter Factors for Neutral Density Filters. . 21 Values of Temperature and EMF for Copper- Constantan Thermocouple . . . . . . . . . . . 55 The Vertical Depolarization Ratio of Carbon Tetrachloride at Various Temperatures . . . . 34 The Value of -lan for Carbon Tetrachloride at Various Temperatures . . . . . . . . . . . 59 Comparison Between Theory and EXperiment for Carbon Tetrachloride. . . . . . . . . . . . . 41 The Vertical Depolarization Ratio of Benzene at Various Temperatures . . . . . . . . . . . 43 The Horizontal Depolarization Ratio of Ben- zene. . . . . . . . . . . . . . . . . . . . . 47 FIGURE 2.1 2.2 2.5 2.4 2.5 2.6 5.1 5.2 4.1 4.2 4.5 4.4 4.5 4.6 5.1 5.2 5.5 5.4 LIST OF FIGURES The Depolarization Apparatus. . . . . . . . . . Alignment of the Laser Beam . . . . . . . . . . The Photometer. . . . . . . . . . . . . . . . . The Temperature Control Unit and Sample Cell. . Plot of Auxiliary Variac Setting vs. Set Point. Schematic of Preamplifier . . . . . . . . . . . Collection Vessel . . . . . . . . . . . . . . . Piece for Rotating Sample Cell. . . . . . . . . Plot of Pv vs. T for CC14 . . . . . . . . . . . Plot of :L/Pv vs. T for cc14 . . . . . . . . . . Plot of -lnPv vs. T for CCl4. . . . . . . . . . Plot of PV vs. T for Benzene. . . . . . . . . . Plot of 1/PV vs. T for Benzene. . . . . . . . . Plot of —lnPV vs. T for Benzene . . . . . . . . Divergence of Pv Data due to Improved Tempera- ture Measurement and Control. . . . . . . . . . Divergence of -lan Data due to Improved .Temperature Measurement and Control . . . . . . Linear Plot of -1an vs. T for CCl4 Constructed from a Limited Number of Data Points. . . . . . Laser Output Power During a Twelve Minute Time Peri-0d. O O O O O O O O O O O O O O O O O O O 0 vi Page 11 15 16 25 26 28 55 57 4O 44 45 46 49 50 52 54 CHAPTER I INTRODUCTION Purpose of this Research Examination of an instrument previously built in this laboratory (1) for the measurement of the depolarization ratio of Rayleigh scattered light as a function of tempera- ture indicated that improvement of several major components of the instrument was needed. Detailed procedures for mak- ing depolarization measurements with this instrument were not available and it was decided that new procedures should be developed and eXplicitly reported. It is hOped that this work will enable future measure- ments of the depolarization in liquids to be made accurately. CHAPTER II THE LIGHT SCATTERING APPARATUS Introduction Figure 2.1 shows a block diagram of the entire apparatus. The laser assembly and the photometer were bolted to a com- posite stone slab 8 ft. x 2 ft. 4 in. x 1 in. This slab rested on tOp of two hollow steel beams 8 ft. x 2i-in. x 2% in. which were supported by a laboratory table. The two beams were used as support for the ends of the stone slab which overhung the table. For ease in discussion the apparatus will be considered in three parts: the light source (laser), the photometer, and the detection system. Light Source Page; A Spectra-Physics Model 120 Stabilite helium-neon laser operating at 652.8 nm was the light source. The specifications listed the output power as greater than five milliwatts and the beam as linearly polarized to better than one part per thousand. The use of this laser mmoumoHHHUmo umquuHo> Hmuflmflc uwHMHHmEm GHLMUOA HmHMHHmEmmHm mammsm HGBOQ mumuumn mnsu uwflamwuaseouozm many “Boga mHQDOUOEHwnu Hmaaoupcou whopmummeu HmEHommcmuu omwum> memono noumuon coflumufiumaom momma m3muom mcflcmflam cowcflm Ucm xomu msumummmé coflumuflumHome mnall.a.m musmflm UH. BA a.m magmas 2m (I 5.4-— was the first major change in the previous system. The laser originally used consisted of three separate components: the plasma tube and two mirrors. Each of these components had to be separately aligned on the optical axis of the ap- paratus. The direction of the laser beam was determined by adjusting the plasma tube and mirrors. The plasma tube was mounted in two tube mounts which could move in both a hori- zontal and a vertical direction, and the mirrors were mounted in gimbal mounts. The direction of the laser beam was thus determined by eight separate adjustments: two for each tube mount and two for each mirror mount. In addition to the alignment difficulties encountered as a result of these eight degrees of freedom, the laser was rf-excited and it was necessary to enclose the plasma tube in an aluminum box to shield the detection electronics from rf-interference. The dc-excited Spectra-Physics laser eliminated this problem and greatly simplified the alignment of the laser beam since the laser could be moved as one unit instead of three. Alignment of the Laser Beam The Optical axis of the apparatus was defined by two apertures on the photometer as shown in Figure 2.2. The first aperture (P1) was the Opening in the iris of the camera shutter (C). This opening is approximately 1/16 in. in diameter. After removing the light trap from its housing, an aluminum plate (P) painted flat black with a 1/8 in. hole cofluomammu xomn mmnspuwmm mcwcflmmw mflxm mumam ESCHEDHM mcwmson mmuu unmwa pass Houpcoo mnsumummamu Hmuusnm mumamo momflm mcflcmfiam ESCHEDHM momma mBmHUm mcwcmaam GOACHQ Ucm xumu mm mmém BA Emmm momma may mo ucchmHH¢|I.m.N musmflm N.m musmgm mm (r through its center, was bolted onto the housing. The hole in this plate defined the second aperture (P2). The Spectra-Physics laser was mounted inside the aluminum box previously used as an rf—shield. The box was supported by a three-point suspension system which mounted on two rack and pinion assemblies (see Figure 2.1). Each end of the box Tuxi a 1/2 in. hole drilled through it to allow the laser beam to emerge. An aluminum cylinder (AP) with a 1/4 in. flange and a 1/8 in. hole drilled through its center was inserted in the front hole. When the laser beam passed through this aperture and the two apertures of the photometer (P1 and P2), it was roughly on the axis of the apparatus. The alignment was completed by observing the back reflection GNU of the laser beam off the front window of the temperature control unit on the aluminum aligning piece (AP). The laser was adjusted with the rack and pinion assemblies and the three-point suSpension system until the back reflection superimposed on the incident beam. Once the laser beam was aligned it was necessary t5 in— sure that the polarization vector of the beam was vertical with respect to the sample. A polarization rotator, which allowed the polarization vector to be rotated through an angle of 560°, was attached to the front of the laser. The technique for aligning the vector was developed by a - colleague (2) and allowed the polarization vector to be . Placed in the vertical position to within 10 minutes (5 milli- radians) of arc . Photometer Introduction A detailed sketch of the photometer is shown in Figure 2.5 with the exception that all electrical connections have been omitted. The laser beam entered the photometer through an opening in the iris (I) of the camera shutter (S). A lens shade (L) 5 in. long and painted flat black inside, was attached to the camera shutter to reduce ambient light. The beam passed into the temperature control unit (TCU), through the sample cell (SC) containing the liquid to be studied, then emerged from the temperature control unit into the light trap (LT). The scattered light was observed at 900 to the incident beam with the photomultiplier tube (PM) after it' passed through a polaroid (1 or 2), the neutral density filters (5 or 4 or both), if required, an interference filter (IF), and a circular polaroid (CP). The temperature control unit and circular polaroid are two major modifications made on the photometer. Temperature Control Unit A cross sectional view of the temperature control unit is shown in Figure 2.4. The unit consists of an asbestos insulated copper cylinder with windows placed every 900. The entrance window and the window through which the scattered light was viewed were made of high quality quartz. The other two windows were Pyrex. The sample cell had inside 10 Figure 2.5.-—The Photometer L S I TCU SC LT IF CP PM I-P-(NNP‘ lens shade camera shutter iris temperature control unit sample cell light trap interference filter circular polaroid photomultiplier tube horizontal polaroid vertical polaroid 50% neutral density filter 10% neutral density filter 11 CP L TCU SC LT L.----__-__.J Figure 2.5 Figure 2.4.-~The Temperature Control Unit and TC G m SC 3’ m "d 2 Sample Cell thermocouple glass tube heating wire sample cell window locating foot leveling screw asbestos 15 TC fins? Fun Figure 2.4 14 dimensions of 50 mm x 50 mm x 60 mm. It rested on a cone of aluminum that had been indexed to hold it in place. The cell could be rotated to enable one of its faces to be positioned perpendicular to the laser beam. The two "feet" of the unit were keyed to fit over two locating pins on the bottom of the photometer. A leveling screw allowed the unit to be tilted slightly about a horizontal axis. The temperature control unit could be heated by four nichrome heating wires. The wires were equal in length and had a resistance of 0.657 ohm per foot. The copper cylinder was first covered with a layer of asbestos for electrical insulation. The four wires were then wrapped on the cylinder, each encircling it eight times. The wires were covered with 1/4 in. to 5/8 in. of asbestos for thermal insulation. The unit was painted flat black on the inside (except for the cone) and on the outside with Krylon ultra flat black enamel spray paint. Only two of the heating wires were needed to provide temperatures up to 950C. One was Operated by an on-off temperature controller. A thermistor in the wall of the copper cylinder acted as the sensor for the controller. The heating voltage supplied by the controller was reduced to 6.5 volts with a transformer. The second heating wire was used as an auxiliary heater." The voltage applied across this wire was adjusted by two variacs in series. The first variac reduced the line voltage and was always set at a reading Of 50. 15 The voltage from the second variac was varied depending on the temperature desired, and was selected so that the con- troller would have approximately equal heating and cooling periods. A list of the settings for the second variac for every 5°C. from 250C. to 95°C. is given in Table 2.1. The data used to determine these values is shown graphically in Figure 2.5. The line voltage was monitored during experi- ments since these settings varied with it. Table 2.1.--Variac Settings for Auxiliary Heater T (0C) Variac Setting 25 O 50 5 55 15 40 17.5 45 21.5 50 25 55 27 6O 29 65 51 7O 55 75 55 80 57 85 58.5 90 4O 95 42.5 16 ucHom pom .m> msflupmm umwum> humaawxsfi mo uOHmll.m.N mnsmflm AOOV unwom umm ooa om om On om om ow om ON a\ _ _ d _ _ _ _ I.oa I.ON Iron |.O¢ \ Itom Outages DEIJEA Axertrxnv 17 The temperature of the sample was measured by a copper- constantan thermocouple placed inside a glass tube that had been immersed in the liquid (see Figure 2.4). An ice bath was used as the reference. The thermocouple emf was found to vary by less than.i 4 microvolts during an eXperiment. Circular Polarizer A circular polarizer was placed between the photomulti- plier tube and the interference filter. It was added because photomultiplier tubes can be sensitive to the polarization of light. Since horizontally and vertically polarized light are converted into circularly polarized light by the circular polarizer, the photomultiplier tube sees only one kind of polarization and this problem is eliminated. Pglaroid Alignment The two polaroids in the photometer were cut from HN-52 polaroid material (1). One had originally been mounted to transmit vertically polarized light and the other to transmit horizontally polarized light. This alignment was checked by first passing the laser beam through a piece of polaroid material that had its transmission axis perpendicular to the polarization vector of the beam. The intensity of the beam passed by the polaroid was observed on a white card with the room lights off. The polaroid was removed and the laser beam allowed to enter the photometer. A pentaprism was placed in the temperature control unit and its orientation 18 adjusted to send the laser beam down the optical train of the photometer. All components were removed from the beam path except for the polaroid that was assumed to have its trans- mission axis horizontal. The intensity passed by this polaroid was observed as before and no difference could be detected. It was concluded that the polaroid was prOperly aligned. No light was observed when both polaroids were in .the beam indicating that their transmission axes were perpendicular. Polaroid Transmission The transmission of the polaroids was measured by pass- ing the laser beam down the Optical train of the photometer with the pentaprism. The interference filter was the only component in the beam. The light was detected by a Lite-Mike with a lens placed in front of it to focus the beam onto the sensing photodiode. TO keep the Lite-Mike on scale a piece of polaroid material held in a rotating mount was placed in the beam ahead of the pentaprism. The laser beam was suf- ficiently attenuated when the polaroid was oriented for maximum transmission. The output of the Lite-Mike was ampli— fied with a Keithley null detector and then recorded with the DVM. The transmission Of the vertically aligned polaroid was determined by first rotating the polarization vector of the laser to the vertical position and recording this intensity. 19 The vertically aligned polaroid was then placed in the beam and the intensity recorded. Ten sets of data were taken. Dark current readings were recorded before and after data collection. The transmission was calculated as the ratio of the intensity with the polaroid in the beam to the in- tensity with it out of the beam after correcting the inten- sities for dark current. The transmission of the horizontally aligned polaroid was measured in a similar manner after rotating the polariza- tion vector of the laser by 900. The results of these experiments are summarized in Table 2.2. .Table 2.2.--Transmission of Polaroids Experiment Vertical Polaroid _Horizontal PolarOid 1 0.725 1 0.014 0.721 1 0.005 2 0.756 :t 0.001 0.756 + 0.001 Calibration of Neutral Density Filters The transmission of the 50% and 10% neutral density filters mounted in the photometer (see Figure 2.5), and the combination of the two, was measured for both vertically and horizontally polarized incident light. .The filter factors for vertically polarized light were measured by first adjusting the polarization vector of laser to the vertical 20 position with the polarization rotator. (The laser beam was attenuated with a 1% neutral density filter and a green wratten gelatin filter to protect the photomultiplier tube, and sent down the optical train of the photometer with the pentaprism. A two megohm anode resistor provided sufficient voltage for detection by the DVM when -450 volts was applied to the photomultiplier. The vertically aligned polaroid, interferencefilter, and circular polaroid were in the beam. The 50% filter was calibrated by first recording the intensity without the filter in the beam and then with it in the beam. These measurements were repeated until ten sets of data had been collected. The dark current was recorded before and after the data collection. The filter factor was calculated as the ratio of the filtered intensity to the unfiltered intensity after correcting the intensities for dark current. The same procedure was followed for the 10% filter and the combination of the two filters. The experiments were repeated with horizontally polarized incident light by rotat- ing the polarization vector 900 and inserting the horizontally aligned polaroid. The results of these experiments are sum- marized in Table 2.5. The transmission of the 50% neutral density filter was also measured with the vertically polarized component of the scattered light from a sample of benzene providing the light source. The eXperimental procedure was essentially 21 Table 2.5.--Filter Factors for Neutral Density Filters Incident Polarization 50% Filter 10% Filter 50% + 10% Vertical 0.554.: 0.001 0.1504 i.0.0004 0.0715 1 0.0005 Horizontal 0.558.: 0.001 0.1502 1.0.0009 0.0725 1 0.0004 the same as that develOped for measuring the depolarization ratio (Chapter III). The results of the two exPeriments run were: EXperiment 1: 0.512 1 0.002 .EXperiment 2: 0.514 1.0.004 Detection A phase-sensitive detection system was used to detect the scattered light. In this type of detection system the signal is produced at a specific frequency and phase. The receiver is tuned to this frequency and phase and rejects any other spectral components accompanying the signal,thus eliminating most of the noise spectrum. .The laser beam was modulated with a mechanical light chOpper Operating at 80 Hz. The chOpper also produced a reference signal of the same frequency and phase as the modulated laser beam. This signal was applied to the refer- ence channel input of a lock-in amplifier. 22 The output Of the photomultiplier tube was amplified by an Analog Devices 142B FET input Operational amplifier. The schematic of the preamplifier is shown in Figure 2.6. The amplified signal was applied to the signal channel input Of the lock-in amplifier. The output of the lock-in amplifier was 1 ma full scale. A 1000 ohm wire wound resistor was placed across the output terminals of the lock-in to provide a voltage which was recorded by a DVM. 25 BDO swamwamammum mo Uflumam£UMIl.m.N musmflm ooa + um 60x50 came h. SO Moa CHAPTER III MEASUREMENT OF THE DEPOLARIZATION RATIO Sample Cell The sample cell was cleaned with a soap solution pre- pared by dissolving a small amount of Tide in about one liter of distilled water. This soap solution was allowed to stand for several hours so that any abrasive particles would settle. It was then decanted into a polyethylene squeeze bottle for use. The sample cell was filled about three-fourths full with the soap solution and then scrubbed inside and out with a camel hair brush that had the bristles trimmed short. The cell was then rinsed well with distilled water followed by five rinsings with conductance water. Purified conductance water was prepared by first passing the laboratory distilled water through a demineralizer and then distilling it through a column 1.5 m long packed with beryl saddles. The cell was dried by placing it in an inverted position inside a vacuum desiccator containing no desiccant, and then pulling a vacuum. The desiccator had been previously purged several times with filtered air. .NO residual water marks were found on the cell when it had dried. The cell was 24 25 always handled with polyethylene hand guards (gloves) to prevent grease from the fingers from smudging the clean surface. Sample Preparation Dust removal is of major concern in preparing samples for light scattering. .Filtration through an ultrafine sintered glass filter under a slight excess pressure of nitrogen was one method tried to remove dust. The filtered sample was collected in a glass weighing bottle which was in a dust free box. The compactness Of the temperature control unit required that the sample cell be placed in it before it was filled. The cell was filled by removing the sample from the weighing bottle with a hypodermic syringe and then trans- ferring it to the cell. This technique proved unsatisfactory probably because the sample was excessively handled. .Dust was removed most effectively by distillation. The sample was distilled with a bantamware still into the col- lection vessel shown in Figure 5.1. The glassware for the distillation was cleaned in an acid cleaner consisting of 5% hydrofluoric acid, 55% nitric acid, 62% distilled water, and a very small amount of Tide. The glassware was rinsed in this cleaning solution, rinsed five times with distilled water, soaked in aqua regia, rinsed ten times with distilled water, rinsed five times with conductance water, and dried. Several milliliters of sample were distilled to purge dust 26 6" ‘.’ X'K— 14/20 standard taper f ' 50 mm tubing 9% 11-— 10" —v-.¢ i? 2 mm Teflon stOpcock ‘5 E}— 10/50 standard taper Figure 5.1.—-Collection Vessel 27 from the still. The collection vessel was rinsed three times with distillate before the final sample was collected. The sample was introduced into the cell by opening the stOpcock, thus reducing the handling of the sample considerably. This method of dust removal and sample transfer proved to be much more convenient as well as effective. Chemical purity was also of concern and Matheson Coleman & Bell "Spectroquality" grade chemicals were used. Sample Cell Alignment A polyethylene glove was worn to prevent glass smears when the sample cell was removed from the desiccator. The cell was covered and examined for dust particles. Any parti- cles found were brushed off with a camel hair brush. The cell was then transferred to the temperature control unit with crucible tongs and located on the indexed cone. The tips of the crucible tongs had Tygon tubing placed over them to grip the cell better and prevent scratching. The cell was rotated with the aluminum piece shown in Figure 5.2 until one face was perpendicular to the incident laser beam. .This aluminum piece also served as the cover for the sample cell while it was being examined for dust particles. The reflection of the laser beam from the cell face was Observed on the aluminum aligning piece described in Chapter II as the cell was ro- tated. The cell was considered aligned when the reflection from the cell face superimposed on the laser beam. Complete 28 l I I I I I I J ‘ 09 32. Figure 5.2.--Piece for Rotating Sample Cell 29 suPerposition could not be obtained and the reflected beam fell just below the hole in the alignment piece. Measurement of Pv The sample was introduced into the sample cell as previously described. The room lights were turned off and the sample was examined for dust particles by viewing it from the top as the laser beam passed through it. Any dust particles present were easily observed as they passed through the beam. The glass tube that held the thermocouple was inserted through the hole in the top of the temperature control unit and any particles of asbestos clinging to it were brushed off. The tOp of the temperature control unit was gently put in place to prevent jarring the sample cell out Of alignment. The temperature control unit was adjusted so that the back reflection of the laser beam from its front window and the sample cell would superimpose on the incident beam. Complete superposition was not possible with one reflection being just above the hole in the aluminum aligning piece and the other just below. The thermocouple was inserted into the glass tube, the probe for the temperature controller placed in the temperature control unit, and the sample compartment of the photometer covered. The temperature controller and the auxiliary heater were set for the desired temperature and the sample allowed to 50 come to thermal equilibrium. This took approximately one hour. .The lock-in amplifier was tuned with an oscilloscope. The two quantities required to calculate the depolari— zation ratio for vertically polarized incident light are the horizontally polarized component of the scattered light, Hv’ and the vertically polarized component of the scattered light, Vv' These two quantities were measured alternately, i.e., first Hv was measured by inserting the polarOid that had its transmission axis horizontal,and then VV was measured by inserting the polardid that had its transmission axis vertical and one or more neutral density filters if necessary. ‘The components for Vv were placed in the path of the scattered light before the horizontal polaroid was removed to prevent too large a signal from entering the detection system. This sequence of data collection was taken ten times for each temperature. Hv and VV were measured in this alternating manner to help reduce the effect of laser in— stability on the data. Noise from the output of the lock-in amplifier was re- duced with the 0.1 sec. RC filter on the instrument.- The DVM was operated with a time base of 10 sec. Two factors had to be considered when recording data from the DVM: .1) After one measurement had been made, the appropriate components were inserted for the next number. 51 The voltage that the DVM recorded during this period had to be discarded since this average contained a time period when no light was reaching the photo- multiplier tube. The next reading (ten seconds later) was recorded. 2) When the temperature controller turned "on" or "off" it injected a pulse into the lock-in amplifier and caused the output voltage recorded by the DVM to be high. If this happened during the measurement of Hv’ this number was discarded and the next one recorded. If this happened during the measurement of Vv’ both Hv and Vv were discarded and the set of measurements repeated. The emf of the thermocouple in the sample was measured before and after each experiment. Ten values of the dark current (the reading the DVM gave when no light was striking .the sample) were recorded before and after the values of Hv and Vv were recorded. The voltage on the photomultiplier tube was always -450 volts. CHAPTER IV DATA Thermocouple Calibration Roeser and Dahl (5) measured temperature-emf relations of a large number of c0pper-constantan thermocouples and found that these did not differ significantly from the rela— tion develOped by Adams (4). They showed further that the deviation of any one thermocouple from Adams' relation was an approximately linear function of temperature. The emf of the thermocouple used in this work was measured in two different experiments at the boiling point of water and found to deviate from Adams' value by only one microvolt. The temperature and emf values derived by Roeser and Dahl from Adams' work were therefore used without modi- fication. These values for the temperature range considered in this work are listed in.Table 4.1. 'To obtain intermediate values of temperature, a linear interpolation was used. Carbon Tetrachloride Data The vertical depolarization ratio of carbon tetra- chloride was measured from 250C. to 71°C. as described in 52 55 Table 4.1.--Values of Temperature and EMF for C0pper- Constantan Thermocouple Temperature (0C) emf (microvolts) 20 787 50 1194 40 1610 50 2054 60 2467 70 2908 80 5556 Chapter III. It was necessary to attenuate the vertically polarized component of the scattered light with both neutral density filters. The average of the two values for this filter factor reported in Table 2.5 was used. The values of Pv were calculated by subtracting the average of the dark current readings taken at the beginning and end of the experi- ment from HV and Vv, dividing this value of Vv by the filter factor, and taking the ratio of H-V and the final value of VV, i.e., . VV-(dark current)aV§] filter factor J (1) P =- H _ " ‘- v ( v dark current)aVg The results of this study are summarized in Table 4.2 where Set Point refers to the temperature selected on the tempera- ture controller, emf refers to the average of the emf value 54 Table 4.2.--The Vertical Depolarization Ratio of Carbon Tetrachloride at Various Temperatures Set Point (0C) emf (uv) T (0C) Pv 25 978 24.7 0.0159 :.0.0005 50 1176 29.5 0.0155 4 0.0005 55 1577 54.4 0.0150": 0.0002 40 1590 59.4 0.0147 4 0.0004 45 1805 44.6 0.0145 i 0.0002 50 2018 49.5 0.0141 4 0.0002 55 2229 54.5 0.0158 r 0.0001 55 2226 54.4 0.0157 r 0.0005 60 2450 59.1 0.0155 r 0.0005 65 2550 61.9 0.0140 i 0.0005 65 2658 65.9 0.0156 4 0.0005 65 2657 65.8 0.0155 i 0.0001 68 2752 66.4 0.0151 4 0.0001 70 2872 69.1 0.0152 i 0.0005 75 2954 71.0 0.0150 r 0.0002 recorded at the beginning and end of the experiment to the nearest microvolt, T refers to the temperature calculated from the emf, and PV refers to the vertical depolarization ratio. The uncertainties in Pv represent the rms deviation. Figure 4.1 shows a plot of Pv vs T. The data reported by Anderson (1) are shown for comparison. Buckingham and Stephen (5) have shown that the vertical depolarization ratio of"a dense fluid of Spherical molecules may be written as 55 0.0174- (3 Previous work I This work 0.0165 0.0156 0.0147 0.0158 0.0129 o.0120*r l 1 L h 1* 1~ 20 50 4o 50 60 70 T (0c) Figure 4.1.--Plot of Pv vs. T for CCl4 c 56 v MR — MRQ\ ioRéT’ 4A + ET 4( MR I + _ 0 V Where MR = molecular refraction of the molecule in the condensed fluid MR0 = molecular refraction of the molecule in the gaseous state R = the gas constant B = the isothermal compressibility 'V = the molar volume T = the absolute temperature The reciprocal of Equation (2) can be taken to give 4. _ 44.1.21. .4 .JR Pv — 5A 5+3AT (5) Thus a plot Of 1/Pv vs. T should be approximately linear depending on the temperature dependence of the quantity B/5A. Figure 4.2 shows a plot of 1/Pv vs. T for carbon tetra- chloride. A stepwise multiple regression analysis (6) was done on the data. It was fitted to a linear equation in temperature with a standard error in the dependent variable of 0.295. When a term in the third power of temperature was added, the standard error was reduced to 0.168. However, both of these standard errors are considerably less than the precision of the data and the linear approximation is therefore adequate at this time. i/pv 57 80.0 r— 75.0 70.0 65.0 62.0 I l l l 20 50 40 50 60 T (0c) Figure 4.2.--Plot of 1/PV vs. T for CCl4 58 Anderson (1) derived the following relation assuming that the quantity A in Equation (2) is independent of temperature: <-gT—1)P [—1—‘5 JIT aT (5:13PM (4) Where PV = the vertical depolarization ratio T = the absolute temperature OT = the coefficient of thermal eXpansion 8 = the isothermal compressibility He further simplified this equation assuming 1) [amp - CV)/OT]P = and 2) (BOT/OT)P = a; to give (é'dngqp = 2 [EBILBL—s] [5 + 0"1'1 (5) The value of -lan determined at various temperatures is given in Table 4.5. wThe uncertainties in -lnPV represent the rms deviation. The data are plotted in Figure 4.5=510ng with Anderson's for comparison. WOOd and Gray (7) gave values of GT and (9%22.)P for O Olan) temperatures of 25°, 40 , and 70°C. The value of ( P was calculated by Equations (4) and (5) and evaluated graphically from Figure 4.5 at these temperatures. The results are summarized in.Table 4.4. 59 Table 4.5.--The Value of -lan for Carbon Tetrachloride at Various Temperatures m J T (0C) ‘lan 24.7 4.14 i 0.02 729.5 4.17 i 0.02 54.4 4.20 i 0.01 59.4 4.22 i 0.02 44.6 4.25 i 0.01 49.5 4.27 i 0.01 54.5 4.29 i 0.01 54.4 4.29 i 0.02 59.1 4.50 i 0.04 61.9 4.27 i 0.02 65.9 4.50 i 0.05 65.8 4.52 i 0.01 66.4 4.54 i 0.01 69.1 4.55 i 0.05 71.0 4.54 i 0.01 40 4.56?" 4.50—- 4020— > o. c H l 4.10.. (3 Previous work I This work 4.00 l l l l J l 20 50 40 50 60 70 80 T (Oc) Figure 4.5.--Plot of -lan vs. T for CC14 41 m.¢ musmwm Eoump Amv coaumsqm Eoumo Adv COAumsvm Eoumn H.d musmwm Eoumm U #H.N d¢.m 0N.m 0500.0 m00«00.0 00WH0.0 mH.m¢m Nm.¢ $5.0 fid.m >500.0 m0NH00.0 m0¢H0.0 md.mfim $0.0 >0.m Nm.m 0500.0 mNNH00.0 mmmd0.0 md.mmm m B L m B I 80 I m an H > OOH x A samv UmOd x A camv QOOfi xofimmflmv ammflmv O m Q AMOV B Hi 4 mpwuoanumuume GOQHMO How ucmEHHmmxm was >HOO£B cmm3umm comwnmmEOOII:¢.¢ manna 42 Benzene Data Most of the benzene data was taken when the instrument was first put into Operation. The sample, except for one data point, was reagent grade and no dust removal was at- tempted. The data are summarized in Table 4.5. .The uncer- tainties in PV and -lan represent the rms deviation. The data is plotted in the same manner as carbon tetrachloride in Figures 4.4, 4.5, and 4.6. Anderson's data are included in Figures 4.4 and 4.6 for comparison. The depolarization ratio for horizontally polarized incident light, Ph, given by P=Jl (6) was measured for benzene at room temperature. The horizon- tally polarized incident beam was obtained by rotating the polarization vector of the laser 900 from the previously determined vertical position with the polarization rotator. The measurements of Vh and Hh were made in the same manner as Vv and Hv were for Pv measurements. The sample was Matheson Coleman & Bell "Spectroquality" with the dust re- moved by distillation. The results of the measurement of Ph are summarized in Table 4.6. The uncertainties in Ph represent the rms deviation. .umsp m>OEmH Op moco OwaawumHO .=>uwamsvouuommm= Hamm a cmeHOO OOmwnumz .x. 45 000.0 0 000.0 0000.0 0 0000.0 000». I--- 4.0500 2000 000.0 0 000.0 0000.0 0 0000.0 0.00 . 0000 00 000.0 0.000.0 0000.0 0 0000.0 0.00 0000 00 000.0 0 000.0 0000.0 0 0000.0 0.00 0000 00 000.0 0 000.0 0000.0 0 0000.0 0.00 0000 00 >000- >0 1060 a I>1V 0&0 1060 0c060 000 mmusumummEmB 050000> um mamucwm mo OHumm cowumuwanomwn Hmofluum> mSBII.m.0 magma 44 0.290" G Previous work I This work 0.280— 0.270— 0.260- 0.2500 0 .240L— 0.250— 0.220.. 0.210- O .200l l I l l l I 20 28 56 44 52 60 68 T (.°c) Figure 4.4.—-Plot of Pv vs. T for Benzene 45 4 .550— 4.450%- 4.550? 4.250‘" > 3‘ H 4.150- 4 .050 r- 5 .950 _ 5.850— 56 44 5.750 28 T (0c) 20 Figure 4.5.--Plot of 1/Pv vs. T for Benzene 46 1.5707 (9 Previous work I This work 1.5405 1.510 " 1.480 — 1.450— 1.420 " 1.560P- 1.550— 1 .500 l J l J l 20 28 56 44 . 52 60 T (0c) Figure 4.6.--Plot of -1an vs. T for Benzene 47 Table 4.6.--The Horizontal Depolarization Ratio of Benzene Experiment T (0C) Ph 1 ~’25 0.987 1.0.009 2 ‘A’25 0.990 1.0.009 CHAPTER V CONCLUSIONS The Spectra-Physics laser, in addition to reducing alignment difficulties, had the added advantage that the polarization rotator c6u1d be attached to it. The polariza- tion rotator facilitates the change between vertically and horizontally polarized incident light making the capacity for measuring PV and Ph at any temperature available. The improvement in temperature measurement and control is shown in Figures 5.1 and 5.2. Figure 5.1 was constructed by requiring the two curves shown in Figure 4.1 to coincide at 25°C. Similarly, Figure 5.2 was constructed by requiring the two curves shown in Figure 4.5 to coincide at 25°C. It is felt that the deviation is the result of an error made in the previous work (1) by equating the set point of the tempera- ture controller to the temperature of the sample. The tempera- ture difference at the high temperature end of these curves is approximately 9.50C. The difference between the set point and control point Of the temperature control unit used in this work can be seen by examination of Table 4.2. 48 pv 0.0174 i— 0.0165 '— 0.0156 '- 0.0147 - 0.0158 .— 0.0129 .— 0.0120 20 Figure 5.1.--Divergence of Pv Data due to Improved 49 “" ‘ Previous work — This work I l l l I 50 40 50 60 70 T (°c) Temperature Measurement and Control -1npv 50 4.56" I 4.50 _- 4.20 .— ..'...__ Previous work This work 4.10 — 4.00 l l L i L 1 20 50 40 50 60 70 80 T (0c) Figure 5.2.--Divergence of -lnPV Data due to Improved Temperature Measurement and Control 51 The previous data (1) were analyzed on the assumption that the natural logarithm of Pv would be a linear function of temperature. However, Figure 4.5 shows that this is clearly not the case for carbon tetrachloride. It is felt that the linear assumption was the result of taking too few data points over the temperature region studied. Figure 5.5 is a plot of -lan vs. T constructed from a limited number of data points. The temperatures chosen were those reported in the previous work. The linear nature of this plot illus- trates the necessity of taking many data points over a temperature range as large as the one covered for carbon tetrachloride. It is for this reason, in addition to the poor quality of the sample, that the benzene data was not analyzed further. Table 4.4 shows that the agreement between the eXperi- mental value of (Riggngp and the theoretical values obtained from Equations (4) and (5) becomes better at lower tempera— tures. It would be interesting to modify the temperature control unit to operate below room temperature and determine if the agreement continues to improve. Table 4.4 also gives an indication of the effect of the approximations that were made in going from Equation (4) to Equation (5). The precision and accuracy of the data was improved by recording the output voltage from the lock-in amplifier with the DVM instead of the strip chart recorder previously em- ployed. It was also much more convenient to record the data 52 4.58F 4.54 ._ ..,. 4.50- 0 4026". ch- > O. c H I 4.22 4- 4.18- 4.14 .— 440 l J L L l J 20 28 56 44 52 60 68 T (0c) Figure 5.5.——Linear Plot Of -lan vs. T for CC14 Constructed from a Limited Number of Data Points 55 digitally. However, still further improvement was indicated by the 1/Pv vs. temperature data for carbon tetrachloride. Figure 5.4 shows how the intensity of the laser used for this study varied with time. With a more stable laser the inten- sity fluctuations could be considerably reduced and the precision in the data would be improved. Cooling the photo~ multiplier Or using a better one to give a higher signal to noise ratio would also improve the precision. Most of the disadvantages of the system were a result of the inconvenience caused by the compactness of the photometer. The electrical connections to the temperature control unit were difficult to make because of the lack of space in the compartment containing the unit (see Figure 2.5). The orientation of the polaroids could not be adjusted unless the photometer was almost completely dismantled. In addition, their present mounting makes any fine adjustments in orientation virtually impossible. These were the primary factors that determined the quality of the alignment check discussed in Chapter II and influenced the decision to assume the transmission of the polaroids to be equal. It is felt that the photometer should be redesigned to alleviate these problems. An additional factor that could have affected the data was that the addition of the circular polaroid was not made until after the carbon tetrachloride data had been taken. The only data taken with it in were the 50% filter factor 54 OO0Mmm 0809 muscflz m>am39 m mcflnsa HOBOm usmuso HmmquI.¢.m mnsmflm Ammuscflfiv mEHB .— . . _ . . . . ,. . . . _ _ . _ _ _ _ . _ _ _ M. In. 0", 0‘. to o o e .0”. F—-— ————————d——-——-—————————-§ O 0 ‘-fl ' --—-—————-—p--—- — ---- --“~---- . mm.o .mm.o 00.0.- 00.0 m¢.o .Nm.o mm.o mm.o (sirun Axexquze) finishequl 55 calibration (Chapter II), the measurement of Pv for benzene at room temperature (Table 4.5), and the measurement of Ph for benzene (Table 4.6). Rerunning the carbon tetrachloride data would serve as a test of the effect of the circular polaroid. Although redesign of the photometer has been discussed, it should be emphasized that the instrument is at present capable of producing useful data. A good deal of data taken while the improvements are being made would be;helpful in determining the effect Of the modifications and the sensitiv- ity of the instrument. LI ST OF REFERENCES R. J. Anderson, "The Depolarization of Rayleigh Scattered Light," Ph.D. Thesis, Michigan State University, 1967. S. J. Gaumer, unpublished thesis, Michigan State University Wm. F. Roeser and Andrew I. Dahl, J. Res. Nat. Bur. Stand. ‘29, 557 (1958). L. H. Adams, International Critical Tables 1, 57 (1926). ~A. D. Buckingham and M. J. Stephen, Trans. Faraday Soc. .§§. 884 (1957). S. J. Gaumer, PROGRAM SMURE, 1970. S. E. WOOd and J. A. Gray, III, J. Am. Chem. Soc. 14, 5729 (1952). 56 APPENDIX APPENDIX EQUIPMENT LIST Laser: Spectra-Physics; Model 120 Stabilite. Polarization rotator: Spectra-Physics; Model 510. Temperature Controller: Yellow Springs Instrument Company, Inc.; Thermistemp Temperature Controller Model 71. Sample cell: Phoenix Precision Instrument Co.; catalog no. T—101. Polaroids: Polaroid Corporation; type HN-52 sheet. Neutral density filters: Baird-Atomic. Interference filter: Spectrolab. Circular polaroid: Polaroid Corporation; type HACP24 x 0.050" (amber) sheet. Photomultiplier tube: RCA 6199. Photomultiplier tube power supply: batteries, -450v to -900v in 90v increments. Lock-in Amplifier: Princeton-Applied Research Corporation; Model JB-5. Light chopper: Princeton Applied Research Corporation; Model BZ-1. Preamplifier: Analog Devices; 1423 FET input Operational amplifier. DVM: Health Universal Digital Instrument. Potentiometer: Electro Scientific Industries; PVB Potenti- ometric Voltmeter-Bridge Model 500. 57 58 Lite-Mike: EG & G, Inc.; Model 560 Lite-Mike with Model 561 Detector Head. Keithley null detector: Keithley Instruments, Inc.; Model 155 Null Detector Microvoltmeter. Demineralizer: Crystalab; Deeminizer. HICHIGQN STRTE UNIV. LIBRQRIES IllII III II 4 312930008 8576