Single Photon Double Slit Experiment

Anastasia Marchenkova
Georgia Institute of Technology, School of Physics, Atlanta, GA 30332 (Dated: June 23, 2011) We have successfully recreated Youngs double slit experiment with both a laser and an incandescent light bulb, providing rapid photon emission and single photon emission, respectively. Both sources show a two slit interference pattern which corroborates the wave nature theory of light. The separation distance between the two slits is calculated to be 0.46 mm 0.01 mm and the width of the slits to be 0.09 mm 0.010 mm, within literature value predictions.

I.

INTRODUCTION TO THE DOUBLE SLIT EXPERIMENT

and destructive interference at 1 d sin = (n + ) 2 where d is the separation of the slits, is the angle from the center of the detector to the slit, n is the order of the fringe, and is the wavelength of the photon [3]. The destructive interference occurs halfway between the bright fringes and the pattern oscillates, decreasing in brightness as the order n of the fringes increases. The intensity I() of the fringes are calculated using I = I0 cos2 d sin sin[a sin / [a sin /
2

In the 17th century, Christian Huygens explained the behavior of light by saying that is acts as a wave. However, it was not until 1801 that Thomas Young performed the double slit experiment and saw bands of light and dark fringes. These bands corresponded to constructive and destructive interference which is a characteristic of waves and provided evidence for Huygens theory. We used Youngs double slit experimental method to study the interference eect of light, by measuring light intensity at dierent parts of the interference bands [1].

II.

THEORY

Youngs double slit experiment in 1801 showed interference patterns in light that could not be explained by the current theories of classical mechanics. However, phenomena like the Compton eect and the photoelectric eects showed that light can also behave like particles. Therefore, a new theory, quantum mechanics, had to be applied; particles such as photons and electrons can behave like particles and waves [1].

where I0 is forward intensity, is the angle normal to the plane of the slits, a is the width of the slit, and d is the distance between slits. At distances which are large as compared to the slit, the Fraunhofer (or far eld) condition can be applied, as well as the small angle approximation, the equation is used to calculate the separation distance d between the slits d= mD y (1)

where m is the order of the fringe, is the wavelength of the light, D is the distance from the double slit to the detector slit, and y is the distance between the center for the main fringe to the end of the mth order fringe. By collecting data about the peaks in voltage or count intensities, formula 1 is used to calculate the separation distance between the two slits. In this experiment, we study the double slit interference pattern from a laser emitting red photons at 670 nm [1] and an incandescent light bulb emitting green photons at 546 nm [5].

III.

FIG. 1: Interference pattern with a two-slit screen. When monochromatic light hits the screen, experiment shows a wavelike interference pattern [2]. Constructive interference, where the fringes are the brightest, occurs at d sin = n

APPARATUS AND EXPERIMENTAL PROCEDURE

The Teach Spin two slit interference apparatus holds the source, double, blocker, and detector slits, the laser, and the incandescent bulb with the interference lter. It is attached to the detector box which holds the photomultiplier tube. The double slit produces the interference pattern, while blocked list can be used to block light from

2 either the left or right slit to view single slit interference fringes. The photomultiplier tube (PMT) is used to record light counts from the incandescent bulb, while a Fluke multimeter records voltage to monitor intensity of the laser diraction patterns. The shutter must be closed while using the laser source. The rapid emission of photons from the laser can damage the PMT. Micrometers along the two slit interference apparatus change the position of the blocker slit and the detector slit. The blocker slit can change the interference pattern to either one slit or two slit. The micrometers are precise to 0.01 mm. By changing the angle of the laser, and adjusting the positions of the double slit, blocker slit, and detector slit, the interference pattern can be aligned. The voltage for the highest intensity band should be at least 0.5 V when the apparatus is properly aligned. The table below summarizes the experimental setup. Highest Intensity Voltage Background voltage Both Slits Blocked (Left) Both Slits Blocked (Right) Left Slit Only Right Slit Only Two Slits 3.153 V 0.010 V 0.00 mm to 5.35 mm 7.80 mm and higher 5.35 mm to 5.84 mm 7.22 mm to 7.80 mm 5.84 mm to 7.22 mm

(a) Scanning the interference pattern of the double slit interference pattern of the laser source.

TABLE I: Micrometer readings for the blocker slit. Values of micrometer readings are accurate to 0.01 mm. These values are used to determine The interference fringes were scanned by moving the detector slit and recording voltage and photon counts for the laser and the incandescent bulb, respectively.
(b) Scanning the interference pattern of the one slit blocked pattern of the laser source.

IV.

DATA AND ANALYSIS IV.1. Laser Source

FIG. 2: Laser source interference patterns. In the double slit, there is an oscillation between high and low values on either side of the maximum intensity peak. The single slit does not produce constructive or destructive interference, so the intensity tapers o on either side from the peak.

The laser was aligned and the interference fringes centered on the detector. To avoid harming the PMT, the shutter is down and voltage is measured using the Fluke multimeter connected to the photodiode OUT on the detector box. The ratio of the maximum intensity for the two slit pattern to the maximum of the one slit pattern was 3.21. Because the maximum intensity was higher for the two slit pattern, it means that there is interference happening from the double slit. The amplitudes overlap and are added together, giving a stronger intensity [3]. The light is more intense coming from the single slit interference pattern at the points where the double slit interference was destructive. The single slit pattern has the highest intensity at the center, and tapers o on either

side. However, the double slit pattern has an intensity that oscillates as cos2 when varies. When d sin = n , 2 n is odd, the intensity goes to 0 due to destructive interference. At these points, the intensity of the single slit pattern is non zero for values of n 1. To calculate the separation of the slits, we use equation 1. Using 670 nm for the wavelength, the rst order fringe, the distance D from the double slit to the blocker slit to be 83.0 cm, we nd the separation distance d of the slits to be 0.46 mm 0.01 mm. The TeachSpin manual says slits come in sizes of 0.35 mm, 0.40 mm, and 0.45 mm [5], so the data ts well with the literature value of the 0.45 mm slit (3.7 % error).

FIG. 3: Scanning the interference fringes for the single photon interference. Graph shows two slit interference pattern, and two one slit patterns. The double slit pattern shows an oscillating intensity around a peak, while the single slit interference slowly tapers o.

IV.2.

Bulb Source

First, the counts were taken for the brightest intensity of the two slit interference. Table II below shows the calculation of the standard deviation of the photon counts. This table shows that the counts dier by about 10 %. Using this standard deviation to apply error to all the results and scanning the double slit interference and two blocked slit interference patterns, Fig. 3 is obtained. We calculate that the count rate is about 106 photons per second, when taking into account the 4% PMT eciency. At the speed of light, a photon spends about 3 ns traveling from the source to the PMT detector. The probability that there is a photon is ight is 0.3 %, and 99.7% of the time there are no photons. There is a very small probability (< 0.01%) that there are two photons released in the apparatus, therefore we have a reliable single photon source [5]. It doesnt make sense to ask what slit the photon went through. As a wave, it goes through both of the slits. If a measurement occurs which identies which slit the photon goes through, the interference pattern collapses [3]. The one slit pattern is more intense at the places where destructive interference occurs and the intensity approaches zero, using the same argument as presented in Section IV.1. When distance d between the centers of the slits is calculated by using equation 1, using 546 nm as the

Run Counts (103 ) 1 124 2 162 3 161 4 132 5 161 6 148 7 134 8 159 9 149 10 162 Average 149 Background 1 Standard Deviation 14

TABLE II: Standard deviation for bulb source single photon counts [4]. Data was taken for one second at the highest intensity. A low background count shows that there are no major light leaks in the apparatus.

wavelength of the green photons from the TeachSpin manual[5], can calculate a separation d of 0.50 mm 0.05 mm and a slit width of 0.09 mm 0.01 mm.

4
V. ERROR VI. CONCLUSION

There is a possible drift of about 0.005 V on the multimeter during laser source measurements. To combat this systematic error, the apparatus was allowed to settle between taking voltage measurements. Touching the device while taking data resulted in the same voltage jumps and therefore care was taken to not shift the apparatus. Handling the interference lter was avoided as much as possible, and the lens itself was never touched. Oils from ngers and microscratches could cause changes in wavelengths and angles of reection, producing aberrant data. Better alignment of the light sources is integral to the success of the experiment. More accurately aligning the laser resulted in more symmetrical and stronger fringes. The rst data run was taken when the peak intensity was 0.504 V. The nal data run, reported here, peaked the intensity at over 3 V.

The wave nature of light was probed using a laser source and an incandescent bulb source. Interference fringes corresponding to constructive and destructive interference are characteristic of the wave nature of light. We found that the separation of the slits was 0.45 mm and the width was 0.09 mm, using both the laser source and bulb source data. We have conrmed the wave nature of the photon and observed constructive and destructive interference patterns in both rapid photon emission as well as single photon emission.

VII.

ACKNOWLEDGEMENTS

Thanks to Professor First and Martin Anquez for providing guidance and the Georgia Institute of Technology for providing experimental equipment.

[1] Georgia Tech Single Photon Double Slit Experiment. [4] Bevington, P.R. and Robinson, D.K. Data Reduction and [2] Youngs Double Slit Experiment Error Analysis for the Physical Sciences McGrawHill, http://www.studyphysics.ca/newnotes/20/unit04_light/ Third Edition. chp1719_light/lesson58.htm [5] Two-Slit Interference, One Photon at a Time TeachSpin, [3] R.P. Feynman R.B. Leighton, and M. Sands. The Feynman Instruments Designed for Teaching. Lectures on Physics