Name: Rudhra Panchal

Student number: 1009345264

LAB – 5
Michelson Interferometer

Course: PHY347H5S

Instructor: Yuchong Li

TA: Liang Haoqiu

Lab partners: Calum and Erin

 ABSTRACT

This experiment utilized a Michelson interferometer to investigate two key optical phenomena. In
the first part, the wavelength of a red laser was determined by translating one mirror and counting the
number of fringe shifts produced. Seven trials were performed, yielding an average wavelength of 678 ±
14 nm which lies within the accepted range for red light. In the second part, a vacuum chamber was
introduced into one arm of the interferometer to measure changes in the refractive index of air as pressure
was reduced from 50 cm Hg to 0 cm Hg. By observing fringe shifts and applying a linear model, the
average change in refractive index was found to be (1.89 ± 0.08) × 10−4, with a rate of (3.1 ± 0.1) ×
10−6 per cm Hg. The extrapolated refractive index of air at 1 atm was 1.0000236, closely matching
standard values and validating the experimental design.
 INTRODUCTION

Interferometry is a powerful optical technique that uses the principle of interference to make
highly precise measurements of wavelengths, distances, refractive indices, and other properties. Among
the most significant and historically impactful interferometric instruments is the Michelson
Interferometer, which has played a central role in validating the wave nature of light and contributed to
groundbreaking experiments. Even the Michelson–Morley experiment that challenged the ether
hypothesis and later paved the way for Einstein's theory of relativity.

Michelson interferometer functions by splitting a coherent beam of monochromatic light typically

from a laser into two perpendicular paths using a beam-splitter. These two beams are then reflected by a
fixed mirror and a movable mirror before recombining at the beam-splitter and forming an interference
pattern on a viewing screen. These fringes arise due to the superposition of light waves (constructive and
destructive interference).

The observed interference pattern depends on the relative phase difference between the two
beams, which in turn is governed by the optical path length they travel. Since light travels to the mirror
and back again, any displacement d m of the movable mirror results in a total optical path change of 2d m If
m complete fringe transitions are observed during this movement, the wavelength λ of the laser can be
determined using:
2d (1)
λ=
m
In addition to measuring wavelength, the interferometer can also detect changes in the index of
refraction n of a medium inserted into one arm of the interferometer. For a known vacuum wavelength λ 0,
the wavelength within a medium is shorter by a factor of the refractive index:
λ0 (2)
λ=
n
When the refractive index of a medium in the beam path changes, this effectively alters the
optical path length causing the interference fringes to shift. By counting the number of fringes N that pass
a reference mark while the pressure is varied, the change in refractive index can be determined using:
N λ0 (3)
ni – nf =
2d
Here, d is the length of the vacuum cell and λ 0 is the vacuum wavelength of the laser.
Furthermore, if the change in pressure is measured during the evacuation of the cell, and the relationship
between pressure and refractive index is linear under low-pressure, and the slope of the n vs. P graph can
be expressed as:
ni– n f N λ0
= (4)
Pi−Pf 2 d (Pi−Pf )
This linear relationship enables the extrapolation of the index of refraction of air at atmospheric
pressure by determining the slope of the line and evaluating it at standard conditions (e.g., 76 cm Hg).
The objectives of this experiment are to determine the wavelength of a red laser by measuring the number
of fringes as the movable mirror is displaced and to evaluate the pressure dependence of the refractive
index of air by observing fringe shifts while the air inside a 3.0 cm vacuum cell is evacuated.
 In all calculations, proper error propagation and consideration of uncertainties, such as
mechanical backlash, mirror alignment, environmental vibrations, and pressure gauge calibration, are
essential to achieving reliable and interpretable results. Through this experiment, we aim to demonstrate
not only the theoretical principles behind interference and wave optics but also the practical challenges
and intricacies of high-precision experimental physics.
PROCEDURE

Refer to the lab manual “Precession [Link]”

DATA ANALYSIS

In the first part of the experiment, the Michelson interferometer was used to determine the
wavelength of the red laser source. This was accomplished by first translating the movable mirror by a
known distance dm and counting the number of fringe transitions m observed during that motion. And
then using equation (1), the wavelength of red light for that particular run was calculated as displayed in
the table – 1 below:

Run # m (fringes) dm (±1µm) λ (nm)

1 20 7 700
2 20 7 700
3 21 7 666
4 20 6 600
5 20 7 700
6 26 9 692
7 29 10 689

Table – 1: In the table above, each trial lists the number of fringe transitions and its corresponding mirror
displacement used to calculate the laser wavelength using the Michelson interferometer.

The average wavelength of red light calculated from the above observations was around
λ = 678.14 ± 13.80 nm
which comes within the range of the literature value of wavelength of red light (620nm – 750nm). Refer
to the Appendix for sample calculations of each corresponding wavelength value in Table-1.

In the second part of the experiment, the interferometer was configured such that one arm passed
through a 2.57 cm vacuum chamber. While maintaining equal arm lengths, the chamber was gradually
evacuated from 50 cm Hg to 0 cm Hg. As the pressure increased, interference fringes shifted due to
changes in the optical path length, allowing us to measure the number of fringe transitions N.

Pi (cm Hg) Pf (cm Hg) N (Fringes) ΔP Δn

50 0 14 50 0.0001847
50 0 15 50 0.0001979
50 0 14 50 0.0001847
Table – 2: The table above summarizes the pressure change, fringe count, and corresponding change in refractive
index for three vacuum chamber trials during the interferometer experiment.
 Using Equation 2, the change in the refractive index of air Δn was calculated for each trial (Refer
to Table – 2 above. The average value of Δn was found to be (1.89±0.08) × 10−4 over a 50 cm Hg
pressure drop. From this, a refractive index change rate of approximately (3.1 ± 0.1) × 10−6 per cm Hg
was determined. This linear relationship was plotted in Graph – 1 below:

Graph – 1: Graph showing the linear relationship between air pressure (cm Hg) and the refractive index of air minus one,
illustrating how refractive index changes with pressure.

DISCUSSIONS AND CONCLUSIONS

For Experiment – 1:

The wavelength of the red laser was determined using Equation (1). When the movable mirror is
displaced by a distance dm, the light beam travels this distance twice, once toward the mirror and once
upon reflection, thus increasing the total optical path difference by 2d m. This doubled path difference is
what causes the observed interference fringes to shift, and therefore must be included in the wavelength
calculation.

To improve the precision and reliability of our measurement, the mirror was moved such that
around 20 or more fringe transitions could be observed per trial. Measuring over multiple fringes reduces
the impact of fringe identification errors and micrometer uncertainty. A single fringe transition
corresponds to a very small mirror displacement; measuring many fringes and then dividing by that
number allows random errors to average out, giving us better results. Additionally, repeating the entire
measurement across multiple runs and averaging the resulting wavelengths helped account for
fluctuations due to setup imperfections, backlash, or environmental disturbances such as table vibration or
air currents. The resulting average wavelength, 678.14 ± 13.80 nm, fits within the accepted range for red
light (620 nm - 750 nm), demonstrating the effectiveness of this method.
 Though this experiment was performed using the Michelson mode, the Fabry-Perot configuration
may have yielded a different value for λ. The Fabry-Perot interferometer produces more closely spaced
fringes and is more sensitive to small optical path changes, but also demands higher alignment precision.
Any discrepancy in measured wavelength between the two modes could arise from misalignment,
multiple reflections, or fringe overlap in the Fabry–Perot setup. Given this, and the reliableness of the
Michelson method for basic wavelength determination, we would place greater confidence in our
Michelson-based result. Measurement accuracy was limited by several factors: the resolution of the
micrometer (±1 µm), possible damage or calibration drift in the micrometer screw, and especially
mechanical backlash which is the slight delay between micrometer movement and mirror response when
changing rotation direction. This was mitigated by ensuring all fringe counting occurred in a single,
continuous counterclockwise motion. When using fringe counting and a known wavelength in the reverse
case, similar limitations would arise: visibility of fringes, ambient vibration, and difficulties in identifying
when a full fringe cycle has passed can all contribute to uncertainty. Nonetheless, careful control of
methodology and averaging allows for a reasonably accurate experimental determination of wavelength
using the Michelson interferometer.

For Experiment – 2:

Using the slope obtained from the linear fit between the refractive index change Δn and pressure
difference ΔP, we extrapolated the refractive index of air at standard atmospheric pressure. With a
measured slope of approximately (3.1 ± 0.1) × 10−6 per cm Hg, the refractive index at 1 atm (76 cm Hg)
was calculated as natm = 1 + (3.1×10−6 × 76) => 1.0000236 ± 0.0000076 which closely aligns with the
literature value of n ≈ 1.0000273 for dry air under standard temperature and pressure. Although a slight
discrepancy exists, it is within a reasonable margin given that this experiment did not control for variables
like humidity or exact temperature. Furthermore, the vacuum gauge reported pressure relative to
atmospheric pressure, so all gauge readings were corrected using P absolute = 76 – Pguage to ensure accurate
pressure measurements.

The experiment assumes a linear relationship between pressure and the refractive index of air,
which is valid for small pressure ranges in ideal gases. To verify this assumption experimentally, one
could collect more data points across a wider pressure range, both below and above atmospheric pressure,
and analyze the resulting n vs. P plot for deviations from linearity. A nonlinear trend, if present, would
become more apparent in extended ranges. Additionally, because the refractive index of gases also
depends on temperature, we can find out this dependency by repeating the experiment at various
controlled temperatures. This would require enclosing the vacuum cell in a temperature-regulated
chamber, introducing air at known temperatures while keeping the pressure constant, and measuring the
fringe shifts as a function of temperature... Such a setup would enable showing the relationship between
refractive index and both pressure and temperature.

Lastly, the first source of uncertainty was fringe counting, particularly during rapid transitions or
low contrast, which could cause over- or underestimation of N. Small vibrations or air currents in the lab
could slightly shift the optical path and alter the fringe pattern mid-measurement. Additionally, the
micrometer used to translate the mirror had a resolution limit (±1 µm) and may have introduced
systematic error if not properly calibrated. Mechanical backlash in the micrometer was mitigated by only
turning it in one direction, but residual play could still have affected precision. To improve the
experiment, we could stabilize the interferometer setup on a vibration-isolated optical table, use a digital
micrometer for finer resolution, and implement a photodetector to automatically record fringe shifts
instead of relying on manual counting. If repeated, we would definitely take more trials for experiment –
2 at varied pressure intervals to strengthen statistical reliability and test the assumption of linearity more
thoroughly because we were able to collect just three values which wasn’t sufficient to give us a good and
reliable relationship.

APPENDIX AND SAMPLE CALCULATIONS

For experiment – 1:

Sample calculation for the wavelength:

2 d 2(7∗10−6 )
λ= m = = 700 * 10^-9 m => 700nm
20

Standard deviation was calculated using the formula:

With the help of the in-built STDEV function in Excel, we were able calculate the S.D. of the lambda
values from table – 1 to be S.D. = 36.5 nm. In order to calculate the uncertainty for the mean value of
lambda, we use the formula:

σ 36.7
Uncertainty = = =13.8 nm
√N √7
For experiment – 2:

Sample calculation for Δn using Equation 3:

N λ0
Δn=
2d

For Trial 1, N=14, λ0 = 660 × 10−9, dm = 2.57 cm

14(678∗10−9 m)
Δn= = 0.0001847
2(0.0257 m)

For standard deviation of Δn values, 0.0001847, 0.0001979, and 0.0001847, using Excel’s built-in
STDEV.S() function, we got σ = 7.6 × 10−6

For the uncertainty in the mean Δn value, we have

 σ 7.6 ×10−6
Uncertainty = = =¿ 4.4 * 10^-6
√N √3
For the Slope of Δn vs ΔP, Since all ΔP=50 cm Hg we calculate Slope = Δn / ΔP = (1.89 × 10−4) / 50

Hence, slope = (3.1 ± 0.1) * 10^-6 per cm Hg.