Stretching Rubber, Stretching Minds:

a polymer physics lab for teaching entropy

Theodore A. Brzinski III∗ and Karen E. Daniels

Department of Physics, NC State University,
2401 Stinson Drive, Raleigh, NC 27705-8202, USA
arXiv:1508.00538v1 [physics.ed-ph] 3 Aug 2015

(Dated: August 4, 2015)

Abstract
Entropy is a difficult concept to teach using real-world examples. Unlike temperature, pressure,
volume, or work, it’s not a quantity which most students encounter in their day-to-day lives. Even
the way entropy is often qualitatively described, as a measure of disorder, is incomplete and can be
misleading. In an effort to address these obstacles, we have developed a simple laboratory activity,
the stretching of an elastic rubber sheet, intended to give students hands-on experience with the
concepts of entropy, temperature and work in both adiabatic and quasistatic processes. We present
two versions of the apparatus: a double-lever system, which may be reproduced with relatively little
cost, and a commercial materials testing system, which provides students experience with scientific
instrumentation that is used in research.

1
I. INTRODUCTION

Rubber has long been proposed as a subject for discussion1–7 , and as a useful demonstra-
tion8–11 or laboratory excercise12–20 in undergraduate thermodynamics curriculum. Rubber
is an attractive choice for undergraduate education because simple and tractable thermody-
namic models, using little more than the First and Second Laws of Thermodynamics, can
be powerfully predictive. Furthermore demonstrations or lab activities utilizing rubber are
generally affordable and tactile, and given student’s familiarity with rubber in their daily
lives (tires, rubber bands, etc.), the subject is both relevant and memorable.
Rubber’s thermodynamic phenomenology is also often counter-intuitive. For example, the
Gough-Joule effect: rubber that is under tension will exhibit a negative thermal expansion
coefficient. Another surprising phenomenon is that a rapidly stretched rubber band will
heat up, while a stretched band that is suddenly released will cool to below the ambient
temperature. These traits challenge student preconceptions, and lead them to think more
deeply about the concepts involved: entropy, temperature, work, heat and the Laws of
Thermodynamics.
These phenomena defy intuition, yet students can explain them with a simple, 1-D model,
with math accessible to undergraduate students. Rubber is composed of polymers: long
chains of repeating molecular structures (monomers) which are free to bend and twist. The
students model these monomers as fixed-length segments which contribute length ±a to the
overall length L of the polymer, as pictured in Fig. 1. This cartoon model allows students
to relate the macrostate of the system to the conformation of the constituent molecules, and
by doing so, understand rubber’s surprising thermomechanics.
While numerous instructional labs have explored the Gough-Joule effect12,13,15–20 , they
universally characterize the system via its statics, and the procedures emphasize the equa-
tion of state rather than the microscopic origins of the phenomenon. Rubber’s thermal
response to adiabatic processes, on the other hand, seems to have previously been reserved
for lecture demonstration11 . We present a laboratory exercise which explores these behaviors
in two parts: First, students measure the change in temperature of rubber in response to
rapid stretching. Second, starting with stretched rubber, students alter the rubber’s tem-
perature, and measure the resulting change in tension. In both cases, students are asked to
compare their observations of the thermal and mechanical dynamics to analytic predictions

2
FIG. 1: A schematic of the simple model employed by students in the analysis of the laboratory
exercises we describe here. A polymer is modeled as a 1-D chain of segments which contribute ±a
to the total polymer length L.

and explain discrepancies in terms of their assumptions in writing the microscopic model.
Two versions of the apparatus are presented: a double lever and a commercial materials
testing system.

II. POLYMER THEORY

While the Gough-Joule effect challenges student expectations, it is simple to explain

in the context of the cartoon model introduced above. As the temperature of rubber is
increased, the molecules wriggle about more energetically, making the more entropic (and
thus shorter) configuration favorable. This explanation can be arrived at analytically as
follows: Assuming the entropy, S, is a natural function of polymer extension L and the
internal energy of the system, U , we can write:

3
    
∂S ∂S
dS = dL + dU (1)
∂L U ∂U L
If we define the work done by the system as hF i dL where F is the tension exerted on the
polymer and the average is over L, and the heat added to the system as T dS, where T is
the rubber’s temperature, we can write the First Law of Thermodynamics as:

dU = T dS + hF i dL (2)

Combining Eqs. (1) and (2), we can write

   
∂S ∂S 1 hF i
dL + dU = dU − dL (3)
∂L U ∂U L T T

Thus, the tension exerted by the rubber is

 
∂S
hF i = −T (4)
∂L U

The polymers which make up rubber are connected at sites called ’crosslinks’21 . As
depicted in Fig. 2, when the rubber is relaxed, the cross-links will be close together and the
polymers free to take any number of folded and looped configurations; However, when the
rubber is stretched, and the crosslink separation approaches the length of the polymers, the
polymers are pulled taut and are able to explore fewer configurations. The configurational
entropy of the system is a measure of the polymers’ freedom of motion, so as the rubber is
stretched and the number of molecular states available becomes smaller, the entropy of the
∂S


material is reduced. Therefore ∂L U
must be negative, so according to Eq. (4), the tension
increases with temperature. Thus, as the temperature is increased, so long as the load on the
∂S


rubber remains constant, it will contract. The precise form of ∂L U
can be calculated from
our cartoon model by counting the number of configurations that can produce a given value
of L, reminiscent of polarization in a 1-D Ising model. For small extensions, by applying
the binomial expansion, students can show that this model predicts the tension is linear in
L, like Hooke’s law.
Rubber’s tendency to change temperature in response to being stretched or released
is another exciting consequence of the molecular origins of rubber’s elasticity. Rapidly
stretching the rubber is an adiabatic process: The system can not exchange heat with
the environment as fast as work is done on it, and the temperature will increase from To

4
FIG. 2: A simple cartoon contrasting a polymeric solid in both a stretched and relaxed state.
Features like loops in the relaxed state are indicative of larger degeneracy.

to To + ∆T . This is because the energy associated with the lost configurational entropy,
Qconf ig = To ∆S assuming ∆T ≪ To , is converted to thermal vibrational energy. The
vibrational energy of each polymer is that of nN linear oscillators, Uvib = nN kT , where n
is the number of atoms that make up each of the N monomers per polymer, so the heat
dUvib
capacity is C = dT
= nN k. This gives the expected change in temperature:

Qconf ig To ∆S
∆T = = (5)
C nN k

Thus, using the same 1-D model, students can calculate the change in the configurational
entropy of a polymer as it is changed from one length to another, and also use that result to
predict the magnitude of the temperature difference. Because polymers consist of a known
number of monomers of known chemical make-up, students can substitute real material
values into their models and obtain dimensional results that can be tested directly against
observation. The monomer in latex, isoprene, has chemical formula C5 H8 . Latex molecules
can have any number of monomers, but for these exercises we suggest students use N = 20.

5
 The derivations above are commonly included in thermal physics curricula, or are assigned
as an exercise4,6,7 . This analysis requires students to employ many of the introductory con-
cepts of thermodynamics and statistical mechanics. Particularly, the students explicitly
utilize the First Law of Thermodynamics, consider heat and work, relate microscopic me-
chanics to bulk behavior, and utilize both the microscopic and bulk definitions of entropy.
Furthermore, students must consider their assumptions, reason about which quantities are
held constant, and interpret the physical meaning of their results. In discussing this anal-
ysis with students, it is helpful to point out that a carbon-carbon bond is approximately
1.5 angstroms in length, yet students will be looking for the effects of these atomic-scale
predictions in a bulk sample measured in centimeters: understanding bulk behaviors as an
ensemble effect of microscopic physics is key to the philosophy statistical mechanics.

III. EXPERIMENTAL APPARATUS

Here we present two experimental apparatuses that have been used successfully for in-
structional laboratory modules for an advanced undergraduate thermal physics course for
sophomores and juniors: a simple double lever, and a commercial Materials Testing System
(MTS). Both apparatuses allow students to conduct the procedures described below, and
separate instructions are provided for each apparatus. While the MTS offers better dynamic
control of the extension, as well as higher resolution measurements, both apparatuses en-
able students to measure both the change in temperature due to adiabatic stretching of the
rubber sheet, as well as to measure the tension as they perturb the rubber’s temperature,
and compare their results to the calculation of Marx et al.4 and Eq. 4 respectively. In our
laboratory exercises, the sample studied is a sheet of silicone elastomer (Theraband brand),
though any such elastic sheet or band would work.
All temperature measurements are taken at the surface of the rubber sheet with a contact-
less IR thermometer. The operation of an IR thermometer is itself an interesting application
of thermodynamics, and can provide an interesting reading assignment in conjunction with
these lab activities22 .

6
 A. Commercial Materials Testing System

Materials Testing Systems are used extensively in research and industry to test the
strength and response of materials under compression or tension. The model we use (Instron
model 5944) is pictured in Fig. 3. One crossbar (labeled as the upper crossbar in the figure)
moves along a rigid rail, while a second is afixed to the apparatus base (labeled as the lower
crossbar in the figure). Mounted to the upper crossbar is a high-resolution load cell. Finally,
attached to both crossbars are clamps which hold the sample. In order to stretch the rubber
in a controlled and uniform manner, a larger clamping surface is required, so the rubber is
first clamped at both ends between pieces of t-slotted aluminum extrusion with applications
of adhesive-backed sandpaper on the clamping surfaces. The MTS then clamps to the ex-
trusion rather than directly to the rubber. Most MTSs have the ability to pre-tension the
sample, increasing the extension of the rubber until some (small) tension is achieved before
re-zeroing both the tension and extension measurements. Pretensioning in this way pulls
the rubber taut without stretching it. Once the material is mounted and pretensioned in
the MTS, the students can stretch the rubber at a prescribed rate to prescribed threshold
values of either the extension or tension.
For many instructors, securing access to an MTS may be difficult compared to the double
lever described below. However, there are a few of significant advantages to utilizing an
MTS. First, while both apparatuses record the tension at a sufficiently high data-rate,
the MTS has the added benefits of simultaneously recording the extension and precisely
controlling either the tension or the extension. This allows students to discuss and interpret
the functional form of the relation between extension and tension while the double lever
does not. Second, because commercial MTSs are primarily used in industrial and research
settings, students are not only being exposed to an exciting, high quality piece of equipment,
but are also learning applicable trade experience as they learn to use the system. Indeed, the
brief experience students had with the MTS in completing this exercise led a small number
to inquire about independent research projects which utilize the system. Finally, because
most MTSs are designed to handle large batches for material testing, the software generally
supports procedure files which prompt users to record any secondary measurements (like
temperature, in the case of this lab), and ensure the correct test parameters and failsafes
are used. Providing students with such files can help to reduce error, and demonstrate the

7
 Instrument
Controls

Load Cell
Upper X-Bar
with clamp
Sample

Lower X-Bar
with clamp

FIG. 3: A picture of the materials testing system (MTS) utilized for these experiments. The key
components are indicated with orange arrows and text labels in the image: the load cell, the upper
and lower crossbars, the sample and the instrument controls.

power of the system to students without risking inadvertent damage or injury due to user
error. That said, we would issue one final note of caution: these instruments present a
crushing hazard if handled improperly, and can be damaged if the appropriate failsafes are
not used. As with any lab equipment, ensure students are informed of how to operate it
safely.

B. Double Lever

A photograph of the double-lever apparatus is provided in Fig. 4. A descriptive video

can be found in the supplemental materials23 . The key feature of this apparatus is a pair
of levers of different lengths whose fulcra share a plane, but are offset by some distance

8
FIG. 4: A photograph of the first version of the double-lever apparatus. The larger lever is traced
in red. As indicated, its fulcrum is along the center of the base-plate. The inner, small lever is
highlighted in green, and rotates about an adjustable fulcrum which is offset from that of the larger,
outer lever. The rubber is mounted between the two levers, and its tension is measured by a Pasco
load cell (highlighted in cyan). For scale, the base plate has dimensions of 12 inches x 36 inches.

d. In Fig. 4, the longer lever is highlighted in red, and the shorter in green. When both
levers are oriented along the direction from the fulcrum of the longer lever towards that of
the smaller lever (as pictured), the distance between the ends of the levers are minimized.
When the orientation of both levers is flipped 180◦ , the distance between the ends of the
levers increases by 2d. The longer lever is anchored at the center of an optical breadboard
which serves as the base-plate, and has threaded holes arrayed in a 1”-square grid. The
shorter lever can be mounted to any of the rows of threaded holes, allowing for adjustments
of d in increments of 1 inch. The shorter lever incorporates a load cell (highlighted in cyan).
The rubber sample is anchored between the load cell and the handle of the larger lever so
that the load cell measures the tension exerted by the rubber. The rubber is adjusted so
that it is held taut, but not stretched. Therefore, when the lever orientations are reversed,
the rubber is stretched by length 2d. The lever enables students to produce considerable
tension in a short time, thus approaching the adiabatic limit.

9
IV. EXPERIMENTAL PROCEDURE

The laboratory procedure is broken into two parts: A series of adiabatic extensions and
releases during which students measure the changes in temperature associated with the
cycle, and a ’sandbox’ component during which the sample is held at a constant extension,
and students are asked to perturb the sample temperature to observe the response and
subsequent relaxation. Before undertaking the activity, students should have derived Eq. 4
and completed the analytic prediction for ∆T during adiabatic stretching as outlined above.
Example course documents for these exercises are included in our supplemental materials23 .

A. Preparation

At the outset of the lab we find it helpful to encourage students to familiarize themselves
with some of the equipment. In particular, the infrared thermometer may require more
nuanced operation than many students realize. Specifically, most such thermometers provide
a laser guide to indicate what surface is being measured. The laser is usually offset by about
an inch from the center of the area of the actual measurement. Furthermore, the area
measured increases as the square of the distance from the thermometer. Thus, we ask the
students to practice by measuring several objects around the room, then to take 5 minutes or
so to familiarize themselves with the operation and anatomy of the experimental apparatus,
particularly if they are using the MTS. Finally, students are provided with a scrap of rubber
and a length of steel wire coil spring, and are asked to qualitatively compare and contrast
the two, and to discuss what sets each item’s spring constant. This final step helps students
contextualize the experiment they are about to perform, and reminds them of the entropic,
rather than atomic origins of the rubber’s elasticity.

B. Adiabatic Stretching

1. Material Testing System

With the rubber in the unstretched position, students measure the length and tempera-
ture of the sample, and zero the force measurement. Next, they rapidly (≥ 1 cm/s) stretch
the rubber to a predetermined length. Students should measure the temperature during the

10
stretching process, and record the largest value observed, as the rubber will immediately be-
gin exchanging heat with the environment, and quickly cool back toward room temperature.
Once it has been characterized in the stretched state, and returned to room temperature,
the procedure is repeated in reverse, returning the rubber to the unstretched state, and gen-
erating a temperature drop. To ensure the results are reproducible, students repeat these
measurements 3-5 times.

2. Double Lever

The procedure is largely the same as for the MTS: With the double lever in the un-
stretched position, students measure the length and temperature of the sample, and zero
the force measurement. Once they are ready to stretch the rubber, the students quickly
swing the levers to the stretched position, and repeat their temperature and length measure-
ments. As above, students should record the largest temperature observed, as the rubber
will immediately begin to cool. Once the rubber’s temperature has returned to roughly
room temperature, students are instructed to quickly return the system to its unstretched
state, and measure the resulting temperature and force drop. To ensure the results are
reproducible, students repeat these measurements 3-5 times.

3. Analysis and discussion.

The procedure for this component is straightforward, regardless of which experimental

design is used, but also provides the most obvious comparison to theory: students have
calculated changes in temperature. Do they match observation or not? Qualitatively, stu-
dents should observe an agreement in the sign of the effect: adiabatic stretching leads to an
increase in temperature, while adiabatic unstretching leads to a temperature drop. Quanti-
tatively, students’ estimates will generally be one or more orders of magnitude greater than
the change measured, depending on their assumptions.
This is a useful opportunity to quell any impulse to label the experiment a failure, and
instead ask students to identify and address the assumptions used in their predictions. Does
using the appropriate number of dimensions improve or reduce the agreement between theory
and observation? In the calculation for Marx et. al4 it is assumed that the polymer goes

11
 Stretching @ 1 cm/s

40

30
F [N]

20

10

0
0 5 10 15 20
ΔL [cm]

FIG. 5: Experimental values of tension plotted vs extension for a piece of rubber stretched adia-
batically on the MTS.

from unconstrained to fully extended. Is that valid? What different assumptions might
improve the theory? Did the entropy of the system change more or less than expected?
We find that few students end up using their measurements of force or length in their
discussion of ∆T unprompted, but with some guidance, students will note that, while the
magnitude of ∆T is comparable whether stretching or unstretching, the change in force
while the system returns to room temperature is not: the tension relaxes in the stretched
state, but remains constant in the unstretched state. This is consistent with Eq. (4), which
∂S


requires that that ∂L U
= 0 for an unstretched polymer. Is it consistent with the students’
model for S (L, U )? Students working with an MTS may notice one more interesting feature:
the tension of the band as a function of extension, as plotted in Fig. 5 doesn’t match the
linear form they derived for Eq. (4). This is because the rubber’s total internal energy and
temperature are both changing during extension.

12
 C. Force and Temperature

1. Material Testing System

Starting with the rubber in the unstretched position, the band is adiabatically stretched
till the tension reaches a preset tension (our students used 49 N) before data recording
starts. An example of the data students will obtain is presented in Fig. 6. After the rapid
stretching the rubber is quasistatically stretched (at a rate of 1 mm/min) until the tension
has increased an additional 1 N. During this step, students using the MTS can directly
measure the extension dependence of the tension, as shown in the first 200 seconds of data
plotted in Fig. 6 . Initially, the band cools, as it is still warm from adiabatic stretching,
so even as the extension increases, the tension decreases. After approximately 100 seconds,
however, the tension approaches linear behavior, as predicted by Eq. (4). Discussing this
behavior with students is a good way to remind them of the relation they’ve derived between
temperature and force.
Once the tension increases by the additional 1 N, the extension is held constant, and
students are asked to breath on the rubber, first with pursed lips (for cool air), then open-
mouthed (for warm air), waiting for the rubber to return to room temperature after each
breath. This provides another opportunity to teach students about adiabatic processes and
call back the prior measurements: blowing with pursed lips generates breath cooler than
open-mouthed breath because as the air passes the students’ lips, it rapidly (adiabatically)
expands.
Next, students are provided with a pistol-grip halogen spotlight, and are instructed to
use it to slowly and evenly heat the entire area of the band. It is important not to overheat
a single spot or the band will snap. Once the students have completed these tasks, they
are encouraged to explore how changing their protocol for heating the rubber changes its
response. While this portion of the lab is open-ended, students should be reminded to make
a note of what they do to the band and when: the data they will use in their reports will
contain only the tension vs time. They must provide information about what was done to
the band and when.

13
 Stretching Stretching
@ 1 cm/s @ 1 mm/min No stretching: length constant (dW=0)
52
Cooling
Breathed on
51 the band thrice: Dashed line:
2 Newton's Law
1 3 of Cooling
50
F [N]

49
Heated
with
Lamp
48

47
0 200 400 600 800
time [s]

FIG. 6: A typical plot of tension vs time as measured on the MTS, labeled to indicate what was
done to the rubber band. After an initial period of quasistatic extension, the band was breathed
on 3 times: once with pursed lips (cool breath, labeled 1), then twice with an open mouth (warm
breath, labeled 2 and 3). The initial, quasistatic behavior can only be measured using the MTS,
not the double lever.

2. Double Lever

Students are instructed to move the rubber to the stretched position and wait for it to
return to room temperature, then re-zero the load cell and start recording the output from
the load cell. From this point, students follow the same procedure as described for constant-
extension on the MTS above: Students are asked to breath on the rubber, then heat it with
a pistol-grip halogen spotlight, and finally to do some open-ended experimentation. A video
demonstration of the procedure with synchronized tension data is provided in supplemental
materials23 . Again, students should be reminded to make a note of what they do to the
band and when.

14
 3. Analysis and discussion.

Because of the open ended nature of the procedure, the analysis undertaken by students
is largely qualitative in nature. They should be able to explain the general shape of the
peaks in the tension, and contrast peaks corresponding to different methods of heating the
band. Some students may even heat the band to a point that the rate of heat loss to the
environment matches the rate at which the lamp heats the band, resulting in a plateau in
the tension.
Students are again able to compare their observations to their analytic predictions. Be-
∂S


cause the band is held at constant extension, ∂L U
remains approximately constant, so
only changes in T will produce changes in F . While students are not able to precisely
measure temperature dynamically in this excercise, they can predict the way in which the
temperature will change as a function of time. The solution to Newton’s Law of Cooling
predicts that an object, starting at temperature To , cools to the ambient temperature, Ta ,
exponentially as T (t) = (To − Ta ) e−at , where a has units of inverse time, and is a material
property of the system. An exponential fit to the final cooling period is plotted as a dashed
line in Fig. 6. The good agreement between the fit and and the data supports the validity
of Eq. (4).

V. STUDENT CHALLENGES

Several aspects of this lab challenge students’ experimental skills, largely because the
structure and premise of the exercise represents a departure from those of the introductory
laboratory exercises to which students are accustomed. These challenges not only provide
useful learning experiences, but also serve to challenge student preconceptions about experi-
mental goals. One of the most common such challenges arises during the adiabatic stretching
portion of the exercise, in large part because students are expected to be critical of their
theoretical predictions. Students often misattribute the disparity between the expected and
measured values for ∆T to mistakes made during the lab, or to limitations of the experi-
mental apparatus (e.g. the practical constraints on extension rate). Though students can
recite the ’hypothesize, test, refine’ model of the Scientific Method by rote, few students are
quick to accept that the theory they are testing is flawed. The default assumption is that

15
the theory is right, and the experiment is wrong. We believe this is a consequence of the
fact that so many introductory laboratory exercises serve simply to confirm theory, or to
measure a physical constant based on the assumed validity of a theory. For many students,
this exercise provides the first opportunity to question the assumptions implicit in their
analytic work as a direct response to contradictory experimental evidence.
The second portion of the lab requires careful note-taking and involves relatively unguided
analysis. Many students will mislabel details of their plots (such as when they remove the
heat source from the rubber), or find that they can’t explain features which they understood
during the execution of the lab due to lapses in record-keeping. This challenge is also one of
the strengths of the exercise: Introductory laboratory instruction often reads like a recipe,
yet experimental research requires creativity and thoughtful analysis. Exercises like this
one, which rely on students to determine which features are of physical interest, and to be
able to discuss the connection to theory, provide students with important insight into how
experimental research is conducted, build experimental skills like note-taking, and serve as
a stepping stone for students who wish to pursue further studies in experimental physics.
Finally, depending on the age and equilibrium length of the band, it may undergo some
irreversible stress relaxation during the second part of the exercise. This feature can be
observed in Fig. 6 between 200 and 500s, where there is an overall decreasing trend in
the tension. If the band is heated enough, as was done in Fig. 6 during the period where
the band was heated with a lamp, this relaxation is accelerated. Students often identify
this feature as ’melting’. Note that after this accelerated relaxation, once the band cools
again the relaxation stops. This behavior can prompt interesting discussions about phase
transitions, hysteresis and annealing.

VI. CONCLUSIONS

We find that the physics of rubber are a useful tool to increase student interest in ther-
modynamics, to extend their experimental skills, and to test their familiarity with the key
concepts of thermodynamics. The adiabatic response of rubber and the Gough-Joule effect
challenge student intuition, yet both phenomena are easily explained with a simple 1-D
model for a polymer. We describe an experiment in which students test this simple model
against the behavior of real rubber. This instructional lab activity has been a part of our

16
undergraduate Thermal Physics course for second- and third-year students for the past two
years.
In the first part of the experiment, students quantitatively compare theoretical predictions
against the measured behavior of adiabatically stretched rubber. The comparison between
experiment and theory requires students to explicitly calculate the conversion of configu-
rational entropy to thermal kinetic entropy. The experimental observation of this process
demonstrates that entropy is a measureable state variable. At the same time, discrepancies
between theoretical prediction and experimental results are used to motivate students to
examine the assumptions made in their analytic work and hypothesize what changes might
produce a more accurate model. Students then measure the response of rubber to pertur-
bations in temperature while at constant extension. This open-ended exercise qualitatively
matches the behavior predicted by the students’ simple analytic model, and challenges stu-
dents to come up with their own experimental or analytic approaches to testing their model.
One example is by demonstrating that the tension of a cooling rubber band is proportional
to Newton’s Law of Cooling.
The experiment is an excellent example of the capacity of thermodynamics to predict
the properties of everyday objects, and of the capacity of experiment to inform theory. It
leads students to explore the First and Second Laws of Thermodynamics, to engage with the
concepts of entropy, work and temperature, to understand adiabatic processes, and build
their comfort level working with simple mechanistic models. The exercise is also a powerful
example of how microscopic physics can manifest in macroscopic systems: while the analysis
is based on the behavior of molecules measured in angstroms, the students observe the effects
in a piece of rubber measured in centimeters. For instructors, it adds a valuable example to
their arsenal, alongside the Ising Model and Ideal Gas. While the hysteretic nature of rubber,
and its tendency to anneal at relatively low temperatures leads to some undesirable practical
considerations, such as sample degradation and confusing contributions to experimental
results, even these provide potentially valuable teaching opportunities. This experiment
has been a valuable component of our undergraduate thermodynamics curriculum, and has
improved the engagement and interest of students in the course.

17
 Acknowledgments

This work was supported by the NCSU Office of Postdoctoral Affairs under the 2013
Professional Development Award, and by the National Science Foundation under Grant No.
DMR-1206808.

∗ Electronic address: [email protected]

1 W. Hirsch, “Rubber Bands, Free Energy, and Le Châtelier’s Principle,” J. Chem. Educ., 79 (2),
200A–200B (2002).
2 B. Smith, “Using Rubber-Elastic Material-Ideal Gas Analogies to Teach Introductory Thermo-
dynamics. Part I: Equations of State,” J. Chem. Educ., 79 (12), 1453–1461 (2002).
3 J. Pellicer, J. A. Manzanares, J. Zúñiga, P. Utrillas, and J. Fernández, “Thermodynamics of
Rubber Elasticity,” J. Chem. Educ., 78 (2), 263–267 (2001).
4 G. Marx, J. Ogborn, and P. Tasnadi, “Rubber as a medium for teaching thermodynamics,”
Eur. J. Phys., 5 (4), 232–237 (2000).
5 L. K. Nash, “Entropy and rubbery elasticity,” J. Chem. Educ., 56 (6), 363–368 (1979).
6 J. P. Sethna, “Statistical mechanics: Entropy, order parameters, and complexity,” in Oxford
Master Series in Physic (Univ. Press, 2006).
7 D. V. Schroeder, An Introduction to Thermal Physics (Addison Wesley Longman, San Francisco,
1999).
8 W. C. Galley, “Mass-Elastic Band Thermodynamics: A Visual Teaching Aid at the Introductory
Level,” J. Chem. Educ., 84 (7), 1147–1149 (2007).
9 M. Dole, “Lecture table experimental demonstration of entropy,” J. Chem. Educ., pp. 754–755
(1977).
10 J. G. Mullen, “Thermodynamics of a simple rubber-band heat engine,” Am. J. Phys., 43 (4),
349–353 (1975).
11 J. B. Brown, “Thermodynamics of a Rubber Band,” Am. J. Phys., 31 (5), 397 (1963).
12 H. A. Ritacco, J. C. Fortunatti, W. Devoto, E. Fernández-Miconi, C. Dominguez, and M. D.
Sanchez, “Thermodynamics Fundamental Equation of a Non-Ideal Rubber Band from Exper-
iments,” J. Chem. Educ., 91 (12), 2195–2199 (2014).

18
13 D. Roundy and M. Rogers, “Exploring the thermodynamics of a rubber band,” Am. J. Phys.,
81 (1), 20–23 (2013).
14 M. Euler, “Hooke’s law and material science projects : exploring energy and,” 57 (1), 57–61
(2008).
15 J. P. Byrne, “Rubber Elasticity: A Simple Method for Measurement of Thermodynamic Prop-
erties,” J. Chem. Educ., 71 (6), 531–533 (2007).
16 J. Mitschele, “Elasticity to Measure Thermodynamic Properties,” J. Chem. Educ., 74 (4), 368
(1997).
17 G. Savarino, “A general physics laboratory investigation of the thermodynamics of a rubber
band,” Am. J. Phys., 59 (2), 141–145 (1991).
18 M. Bader, “Rubber elasticity-a physical chemistry experiment,” J. Chem. Educ., 58 (3), 285
(1981).
19 J. E. Mark, “Rubber elasticity,” J. Chem. Educ., 58 (11), 898–903 (1981).
20 H. B. Carroll, “Rubber Band Experiment in Thermodynamics,” Am. J. Phys., 31 (10), 808
(1963).
21 R. A. Jones, Soft Condensed Matter (Oxford University Press, 2002).
22 W. R. Barron, “Principles of Infrared Thermometry,” Sensors Mag., 9, Z–59–Z–62 (1992).
23 Supplemental Materials available at http://nile.physics.ncsu.edu/cgi-bin/webpage/
Downloads/.

19