Experiment 5: Spectra of

Polyenes and Nanoparticles

Julia Carlstad
ID# 1230835
Due: 5/12/15
Chem 461

Introduction:
Spectroscopy provides chemists with an effective way to analyze properties of particles
that are otherwise difficult to study using macroscopic techniques. With the development of
nanoparticles that have been found to have energetic and conductive properties, spectroscopic
techniques have become vital to further the scientific analyses of these semiconductors. One such
property that can be monitored using spectroscopy is the size of a particle, which can then be
applied to the quantum models of particle-in-a-box, the foundation of quantum mechanical
analysis of electron probabilities. Understanding the accuracy of the models allows us to better
predict which materials could serve as semiconductors for use in technologies and energy
sources. Electronic transitions in particular can be monitored using UV-Vis spectroscopy. The
measured wavelength of light absorbed is then used to calculate the energy in Joules according to
the equation:
E=

hc

(equation 1)

where h is Plancks constant (6.626E-34 Js), c is the speed of light (2.998E8 m/s), and is the
wavelength in nanometers.
This energy can then be used in further calculations to determine the length of the box,
from the particle-in-a-box model. This calculation can be done according to the equation:
E n+1,n=

( 2n+ 1) h2
8 m e L2

(equation 2)

where n is the quantum number, me is the mass of an electron (9.109E-31 kg), and L is the length
of the box in meters. The length of the box indicates the region in which the electronic
distribution can be found and strong models of this region are crucial in predicting the energies
observed.
The simple model utilized in the understanding of equation 2 can be expanded to a
particle-in-a-sphere in order to better describe nanoparticles and quantum dots, which are of
more interest, scientifically. The radii can be given by particle manufacturer or calculated if the
region of interest falls between 330nm < < 370nm using:
1240
294.0 1.09
=3.301+ 2
half
d
d

(equation 3)

where

half

is the wavelength that falls at half of the absorbance of the maximum wavelength

in nanometers and d is the particle diameter in Angstroms. Once the radii are determined they
can then be used to determine the energy of the transition according to the equation:
E=E gap +

2 2 1 2 2 1
1.8 e 2
+

2 a2 me 2 a2 mh 4 0 a

where E is the energy of the lowest energy absorption transition,

(equation 4)
E gap is the band gap of the

semiconductor, is Plancks constant (h) divided by 2, a is the radius of the radius of the

sphere, m e is the effective electron mass, m h is the effective hole mass, e

free charge (1.602E-19 C),

is the electron

is the dielectric constant of the semiconductor and 0

is the

dielectric constant of free space (8.854E-12 C2 J-1 m-1). Essentially, this equation involves
summing the energy difference between the valence and conduction bands, the electron
confinement energy, and confinement energy of the hole, then subtracting the coulomb attraction
of the hole and the electron.
Through UV-Vis spectroscopic analysis the size of the box in the particle-in-a-box model
(equation 2) can be tested using the data from the phenyl polyenes to determine the accuracy of
the determination of the length of the box. Analysis of the nanoparticles can test the theoretical
model of equation four that indicates that the energy of the excited particle is a function of the
size of the particle. These theories will be tested through the comparison of this model to
energies determined experimentally through recorded spectra.
Experimental:
The experimental handout was followed closely during the laboratory period1. The only
significant difference was the analysis of CdSe crystals which was performed using a different
set of manufactured crystals. These were taken in a reference solution of toluene and information
about their size and excitation peaks were provided by the instructor for use in data analysis2.
Data Analysis:
Diphenyl Polyenes:
Three diphenyl polyene samples of 1,4 diphenyl-1,3-butadiene, 1,6 diphenyl-1,3,5-hexatriene,
and 1,8 diphenyl-1,3,5,7-octatetraene were provided in class and spectra were taken with a
reference solution of cyclohexane between 300-420nm. The results are shown in figure 1.

Figure 1: UV-Vis Spectra of Diphenyl Polyenes. In blue is 1,4 diphenyl-1,3-butadiene, in green is 1,6
diphenyl-1,3,5-hexatriene, and in red is 1,8 diphenyl-1,3,5,7 octatetraene

The maximum wavelength of the lowest energy transition was determined while in lab using the
Spectra Suite program to identify the highest wavelength peak location. These results are
summarized in table 1. The error in these values is estimated based on the accuracy of the
instrument.
Table 1: Diphenyl Polyene Peak Wavelengths
Peak Wavelength
Sample Name
0.005 (nm)
1,4 Diphenyl-1,3butadiene
346.33
1,6 Diphenyl-1,3,5hexatriene
371.74
1,8 Diphenyl-1,3,5,7octatetraene
395.72

Equations 1 and 2 can then be rearranged to determine the measured bond length. A sample
calculation is performed below with 1,4 diphenyl-1,3-butadiene and the data is summarized in
table 2. The error is propagated through these calculations using Gausss error formulas
including R =

R
, 2A + B= 2A + 2B , CA =C A and 2AB= 2A + 2B . These show that
R
3

the data has a reasonably high confidence limit, but that due to error the number of significant
figures that can be reported has decreased with calculation3.
L=
L=

(2 n+1 ) h
8 me c

5 (6.626E-34 Js )((346.33 0.005) E9 m)

m
8(9.109E-31 kg)(2.998E8 )
s

L=7.24 0.02
The theoretical predicted bond lengths were determined using the average bond length of the
carbon-carbon bond (1.39 ) and multiplying by the number of carbon-carbon bonds in the
chain.
Table 2: Diphenyl Polyene Bond Length
Predicted Bond
Sample Name
Length ()
1,4 Diphenyl-1,3butadiene
6.95
1,6 Diphenyl-1,3,5hexatriene
9.73
1,8 Diphenyl-1,3,5,7octatetraene
12.51

Measured Bond Length

0.02 ()
7.24
8.88
10.39

Taking into account the error in the measured bond length the data does not align with the
predicted bond lengths which prevent conclusive analysis of the measured data in comparison
with the model. Even with this, it appears that the bond length increases with the increase in the
length of the molecule which matches with the predicted bond lengths using the energy
eigenvalues. The model used to calculate the predicted bond lengths has limitations in that the
conjugated bonding structure of the diphenyl polyenes has high electron delocalization which
changes the actual bond lengths. This conjugation extends to the phenyl rings, which could be
included in the length of the box seen by the electrons, while the model assumes that the length
of the quantum box is only inclusive of the length of the carbon-carbon chain. In addition, the
orbitals of the atoms extend beyond the distances between the center of mass of each atom,
which means that there is a probability that the electrons are located at a further distance than the
model would predict. These factors could influences the discrepancies between the model and
the experimentally determined bond lengths.

CdSe Nanoparticles:
The Cadmium Selenide nanoparticles were provided by the NNCrystal US Corporation and were
prepared with toluene as the solvent. The spectra of these samples were taken at room
temperature with a reference solution of toluene. Figure 2 shows all spectra plotted from 350700nm.

Figure 2: Uv-Vis spectra of all CdSe nanoparticles. These are labelled according to the producers chart
which was provided in the lab2. Sample Green is plotted in green, sample Green-Yellow is plotted in
yellow, sample Orange is plotted in orange, sample Orange-Red is plotted in red, sample Red is plotted in
burnt orange, and sample Yellow is plotted in blue.

For each sample plotted in figure 2 the maximum wavelength (max ) was determined for the
lowest energy absorption transition, which can be identified on the figure as the region between
500-600nm. This was performed using the peakfinder function in Matlab after narrowing the
data set to the desired region. An example identification of max is shown in figure 3.

max

Abs

Figure 3: Zoomed in view of the CdSe Green UV-Vis spectra to show identification of max.

The maximum wavelength was found for each CdSe sample using a similar method. This
technique proved less effective for samples Orange, Red, and Orange-Red, as their absorption
peaks were less identifiable. This introduced larger error in this determination, and required more
hands-on analysis so as to not obtain a value that is actually in the pre-peak region. For this
reason the reported error in these samples is larger. Overall there is larger error than that of the
instrument for the CdSe nanoparticles because there is not a well-defined maximum wavelength.
As can be seen in figure 3 there is a larger region that has an absorbance that is similar to the
peak absorbance. This introduces larger error because there is a larger region that could possibly
be attributed to the maximum wavelength. The results of the maximum wavelength
determinations are summarized in table 3.
Table 3: CdSe Nanoparticle Peak
Wavelength
Peak Wavelength
CdSe Name (nm)
Green
502.5 0.2
Green-Yellow 517.0 0.2
Yellow
542.3 0.2
Orange
560.5 0.3
Orange-Red
585.8 0.3
Red
598.0 0.3
The energy of the lowest energy absorption can be calculated using these measured wavelength
values. This calculation was performed using equation 1 and an example calculation is shown

below using sample Green. Error was propagated through these calculations using Gausss
formulas with the relative error of the measurements3.
( 6.626 E34 Js)(2.998E8
E=

m
)
s

(502.5 0.2) E9 m

E=(3.957 0.001) E19 J

E=(3.957 0.001) E19 J

1 eV
( 1.602E-19
J)

E=2.4677 0.0002 eV

The data from these calculations for all samples is summarized in table 4 below. The error that is
carried through in these calculations is relatively small, which shows the precision of the
instrument used to measure the wavelengths.
Table 4: CdSe Nanoparticle
Measured Energy
Eexciton measured
CdSe Name (eV)
Green
2.4677 0.0002
Green-Yellow 2.3984 0.0009
Yellow
2.866 0.001
Orange
2.2123 0.001
Orange-Red
2.1168 0.001
Red
2.0736 0.001
Using the identified max values the diameters of each nanoparticle were determined using the
table provided in lab from the manufacturer. These were then multiplied by to determine the
radii which could then be used to determine the theoretical exciton energy. An example
calculation of the exciton energy using sample Green is shown below using equation 4. For the
CdSe particles the values required for the equation are

E gap=1.74 eV =2.787 E19 J ,

m e =0.13(me ) , m h=0.45( m e ) , and =9 .

2

E=2.787E-19 J +

1

1
1.8 e
+

2
2
m
m
4

((1.18
0.03) E9m)
2(( 1.18 0.03)E9 m) e 2(( 1.18 0.03) E9 m) h
0

E=2.787E-19 J + [ ( 3.35957 0.09 ) E19 J ] + [ ( 9.70545 0.2 ) E20 J ] [(3.92616 0.09) E20 J ]
E= ((6.7245 0.2) E19 J )

1 eV
( 1.602E-19
J)

E=4.198 0.1 eV

This calculation was performed with all samples and the results are summarized in table 5 below.
The error in the diameter and radius comes from the range in size that is given by the
manufacturer and this propagates through to the exciton energy2.
Table 5: CdSe Particle Theoretical Energy
Diameter
Radius
Eexciton
CdSe Name (nm)2
(nm)
Theoretical (eV)
Green
2.35 0.05
1.18 0.03
4.2 0.1
Green-Yellow 2.5 0.1
1.25 0.05
3.9 0.2
Yellow
2.8 0.2
1.4 0.1
3.4 0.3
Orange
3.25 0.05
1.63 0.02
2.98 0.04
Orange-Red
3.75 0.03
1.88 0.02
2.6476 0.01
Red
4.5 0.2
2.25 0.1
2.3489 0.05
The theoretical and experimentally determined exciton energies can be compared to analyze the
alignment of the data with the model. This analysis is performed and plotted in figure 4. Error on
the graph is displayed in accordance with the error shown in tables 4 and 5.

Figure 4: The blue circles represent the experimentally determined energies while the blue line represents
the energies determined using the model. The error is shown in the x and y directions on both the
theoretical and experimental data points. This error can be seen numerically in tables 4 and 5 and varies
between each data point.

It can be seen from figure 4 that some of the data points have a larger relative error, which comes
from the differences in error in the radii. The relative error in the plotted experimental energies is
8

Q5

very small. From this plot one can observe that the theoretical and experimental data do not
align. It seems that if the plots were extended to the y-intercept they may come together, but they
have drastically different slopes with the theoretical values having a steeper slope than the
experimental. There does appear to be dependence on 1/R2 in the data, as expected theoretically,
because the experimental values are consistently increasing with increasing values of 1/R2, but
the discrepancies seem to come in the coefficients of these variables. It could be that the model
does not accurately describe the energy because of factors that are not taken into account, such as
how the effective masses and static dielectric constant are influenced by the diameter of the
particle. Further research should be conducted with more sensitive instruments to lessen the error
in the determination of the maximum wavelength to obtain more conclusive information about
the nature of the experimental trend.
ZnO Nanoparticles:
Figure 5 shows spectra of ZnO nanoparticles at room temperature and figure 6 shows spectra of
ZnO nanoparticles at 50C. From this figure it can be seen that the lowest energy absorption
increases in wavelength with time.

Figure 5: UV-vis spectra of all samples taken at room temperature are plotted and are identified by the
time at which the spectrum was recorded. The spectrum taken at 0 minutes is in yellow, 1 minute is red, 2
minutes is green, 4 minutes is black, 8 minutes is pink, 16 minutes is cyan, and 32 minutes is blue.

Figure 6: UV-Vis spectra of all samples taken at 50C identified by the time at which the sample was
removed from the heat bath and quenched. Bright green represents the sample taken at 1 minute, the
sample taken at 2 minutes is in pink, 4 minutes in cyan, 8 minutes in red, 16 minutes in blue, 30 minutes
in purple, 40 minutes in orange, and 120 minutes in dark green.

The maximum wavelength was determined using a similar method as was used to determine the
maximum wavelength of the CdSe nanoparticles. This was performed using the peakfinder
function in Matlab. An example determination is shown in figure 7 with the ZnO sample at 0
minutes. The resulting wavelengths and intensities of all samples are summarized in table 6
below, and are reported with the systematic error intrinsic in the instrument. The error in this data
comes from the noise that can be seen on the plotted spectra in figures 5 and 6, which makes it
more difficult to determine exactly where the peak wavelength is because of the many relative
minima and maxima within the data.

10

max

Abs

Figure 7: Determination of max shown using the data of ZnO toom temperature time 0. The length of the
black line is the absorbance of this peak. Where the absorbance is maximized identifies the peak
wavelength.

Table 6: ZnO Peak Wavelengths

Room
Peak Wavelength
Peak Absorbance
Temperature
0.005 (nm)
0.0005
Sample Time
(min)
0
283.28
0.864
1
285.00
0.893
2
288.42
0.899
4
287.99
0.917
8
289.91
0.88
16
298.44
0.857
32
302.28
0.93
50C Sample
Time (min)
1
317.80
0.603
2
320.76
0.605
4
324.16
0.545
8
327.80
1.063
16
329.40
0.534
30
333.00
0.517
60
335.60
0.574
120
340.40
0.813
The experimental values of Eexciton can then be calculated from the peak wavelength using
equation 1 in a similar manner as was done for the CdSe nanoparticles. These values are
11

summarized in table 7. Error was propagated in the same way as for the CdSe energies, using
Gausss formulas3. It can be seen that the relative error in these measurements is much less than
for the CdSe particles because of the higher confidence in the measured maximum wavelengths.
This allows for the propagation of more significant figures into the energy values.
Table 7: ZnO Measured Eexciton
Room Temperature
Eexciton measured
Sample Time (min)
(eV)
0
4.3773 0.00008
1
4.3509 0.00008
2
4.2993 0.00007
4
4.3057 0.00007
8
4.2772 0.00007
16
4.1549 0.00007
32
4.1021 0.00007
50C Sample Time
(min)
1
3.9018 0.00006
2
3.8658 0.00006
4
3.8253 0.00006
8
3.7828 0.00006
16
3.7644 0.00006
30
3.7237 0.00006
60
3.6949 0.00006
120
3.6428 0.00005
The data of max was then used to determine the half by calculating the half peak absorbance,
taking 1/2 of the maximum peak absorbance, and then locating this half peak absorbance in the
data set. This representation is shown in figure 8, where the location of half can be traced over
from half of the total absorbance value that is shown in figure 7. This process was performed for
all samples and is summarized in table 8. The relative error in the half peak wavelengths has
increased in relation to the maximum wavelengths because there is a larger uncertainty since the
exact half absorbance values cannot always be found in the data collected. This requires more
rounding and estimation, which decreases the confidence in the values obtained. For this reason
more error has been introduced into the recorded data.

12

max

Abs/2
half
Abs/2

Figure 8: Determination of half wavelength using the previously determined values of maximum
wavelength.

Table 8: ZnO Half Wavelengths

Room
Half Peak
Half Wavelength
Temperature
Absorbance
0.01 (nm)
Sample Time
0.0003
(min)
0
0.4320
298.2
1
0.4465
300.6
2
0.4495
302.3
4
0.4585
304.2
8
0.4400
308.0
16
0.4285
312.9
32
0.4650
316.7
50C Sample
Time (min)
1
0.3015
331.6
2
0.3025
334.5
4
0.2725
340.6
8
0.5315
358.3
16
0.2670
348.7
30
0.2585
350.8
60
0.2870
353.5
120
0.4065
359.6
The values of the half wavelengths can be used with equation 3 to determine the diameter of the
particles. The radius, which is required for calculations of Eexciton, can then be calculated by
13

taking half of the diameter. Error is propagated through these calculations from the error in the
maximum wavelength. The model incites larger error in the diameters because it does not cover
the full spectrum of the wavelengths gathered experimentally. The model is said to fit 330nm <
< 370nm, but some of the wavelengths were found to be as low as 298.2 nm which lowers the
confidence in the model. The data from all trials is summarized below in table 9.
An example calculation is performed using the data from the sample taken at 0 minutes.
1240
294.0 1.09
=3.301+ 2
298.23 0.01
d
d
4.15786 0.0001=3.301+

294.0 1.09

2
d
d

d=19.17 0.0006
a=9.585 0.0003
Table 9: ZnO Radii
Room Temperature Diameter
Radius ()
Sample Time (min) ()
0
19.170 9.585 0.0003
0.0006
1
19.558 9.779 0.0004
0.0007
2
19.849 9.925 0.0004
0.0007
4
20.189
10.095
0.0007
0.0004
8
20.909
10.455
0.0007
0.0004
16
21.917
10.959
0.0007
0.0004
32
22.788
11.394
0.0007
0.0004
50C Sample Time
(min)
27.153
13.5765
1
0.0008
0.0004
28.293
14.1464
2
0.0008
0.0004
31.091
15.5455
4
0.0009
0.0005
23.2355
8 46.471 0.001
0.0005
16
36.116
18.0580
14

0.001
0.0005
37.837
18.9185
30
0.001
0.0005
40.424
20.2120
60
0.001
0.0005
48.499
24.2495
120
0.001
0.0005
The radius can then be used with equation 4 to calculate the theoretical value of Eexciton in a
similar manner as was done for the CdSe nanoparticles. Error propagation was performed in the
same way as for the CdSe nanoparticles. These calculations are summarized in table 10. It can be
seen that the relative error in the theoretical calculations is small, which allows four significant
figures to be reported.
Table 10: ZnO Theoretical E exciton
Room
Eexciton Theoretical
Temperature
(eV)
Sample Time
(min)
0
5.287 0.0002
1
5.198 0.0002
2
5.136 0.0002
4
5.066 0.0002
8
4.930 0.0002
16
4.763 0.0002
32
4.638 0.0002
50C Sample Time
(min)
1
4.1887 0.0001
2
4.1065 0.0001
4
3.9445 0.0001
8
3.544 0.00008
16
3.7495 0.0001
30
3.7016 0.0001
60
3.6421 0.0001
120
3.5201 0.00007
The comparison between experimental and theoretical energies is plotted with error in figure 9.
From this it can be seen that the model does not align with the experimental data, in that the plots
have different slopes and y-intercepts. For this reason further analysis of the model must be
performed. The energy appears to have dependence on the radius experimentally, because the
energy increases linearly with respect to 1/R2, but because of the disagreement with the
theoretical model there could be a larger dependence on the size of the radius in the model,
15

requiring adjustments to lessen the increase of energy with respect to 1/R2. The y-intercept of the
experimental and theoretical data for the ZnO particles also do not align which could indicate
that the band gap energy used is not accurate, and could require further verification.

Figure 9: Plot of the experimentally and theoretically determined exciton energy. Error in the x-direction
is plotted according to table 9 for the radii. Blue circles represent the experimentally determined values
with error in the y-direction plotted corresponding to the error in table 7 for energy. The green line
represents the theoretically determined values with error in the y-direction plotted according to table 10.

The room temperature sample was tested the following lab period and the resulting spectra are
plotted in figure 10. It can be seen from the plot and through the use of the model, that the
overnight sample has continued to grow in radius and that the peak is no longer as defined. It is
possible that the maximum wavelength has stopped increasing, but a time-based analysis would
need to be performed to ensure that the wavelength, and thus radius is no longer changing.

16

Figure 10: Plot of the ZnO UV-Vis spectra taken at room temperature on the day the samples were made
and then after sitting for the weekend. The red plot is the spectrum taken on the day of reaction and the
blue plot is the spectrum taken the following lab period.

Q4

The overnight comparison was also performed with the samples that were heated to 50C and
sampled at 1min, 16 min, and 120 min and these spectra are plotted in figure 11. In analysis of
the spectra one can look for an equilibrium position, at which the sample is no longer growing
because the rates of formation are equivalent to the rates of dissolution of the particle back into
solution. This means that no change in the size of the particle should be observed.
Spectroscopically, one would look for the wavelength of the particle no longer increasing with
time. Analysis of figure eleven shows that there were differences in the maximum wavelength of
the three initial spectra taken on day 1, but it appears that the spectra have reached an
equilibrium point by the second day of sampling since all of the maximum wavelengths are
within proximity of each other. This could indicate a possible maximum size that the particles
can grow to, or a maximum energy that can be absorbed by the particle. In order to draw more
concrete conclusions about this equilibrium hypothesis analysis with a greater number of
overnight samples is required and this data should be processed numerically. It also would be
interesting to investigate the time at which the particles reach the equilibrium size, which could
be done using time-based analysis. Long term studies should be done to ensure that the particles
are truly in equilibrium and are not still growing at slower rates.
17

Figure 11: UV-Vis spectra of the overnight samples plotted against the corresponding sample from the
first analysis. The sample taken at 1 min is plotted in burnt orange and the overnight sample 1 is plotted in
light orange. The sample taken at 16 minutes is plotted is dark green and the overnight sample 5 is plotted
in green. The sample taken at 120 minutes is plotted in blue and the overnight sample 8 is plotted in cyan.

Conclusion:
The relative bond lengths of diphenyl polyenes can be determined from the spectra as the
wavelength increases with increasing bond length. There is significant discrepancy between the
theoretical particle-in-a-box model and the experimental values which could indicate factors that
are not being considered in the model such as the electron delocalization due to conjugation that
extends the length of the box seen by the electrons.
The large error in cadmium selenide maximum wavelengths limits conclusions that can
be drawn from the data. There are large discrepancies in the experimental and theoretical trends,
particularly in the slopes of the energy to radius comparison, which could indicate further radial
dependence that is not taken into account in the model.
The ZnO nanoparticles are found to have increasing energy with increasing growth time
and larger radii, but the model does not accurately predict the trends observed. The band gap
energy of the experiment seems to differ from the model predictions as does the slope. Similar to
the CdSe particles, it is possible that there are more terms that also depend on the size of the
18

radius. These predictions should be further analyzed to determine if the experimental trends
observed are repeatable, in which case the model could be revised.
Equilibrium particle size is achieved through overnight growth as the wavelengths of
samples grown to different original sizes have all equalized. Time-based experiments should be
performed to determine when this equilibrium point is reached and to confirm that the particles
are not still growing slowly.
References:
1

Fugimoto, Bryant; Reid, Philip; Ginger, David; Gamelin, Daniel. Experiment 5-Spectra of
Polyenes and Nanoparticles. Seattle: University of Washington, 2012.
2

Cadmium Selenide Quantum Dots. Fayetteville, AR: NNCrystal US Corporation, n.d.

Robinson, Bruce. Error Analysis for Laboratory Science. Seattle: University of Washington.
n.d.
Questions:
1. Quartz is transparent in the visible because the wavelength that is absorbed by this
transition corresponds to 139.33 nm when calculated using equation 1. This is in the UV
region of the electromagnetic spectrum, which means that it does not reflect any of the
visible region of the spectrum, meaning that it is transparent. The longest wavelength of
light that I would expect quartz to absorb due to electronic absorption is 139.33 nm.
2. The band gap of CdSe is 1.74 eV and that of ZnO is 3.4 eV. The value of kT at room
temperature (298.15 K) is 4.1166E-21 J=0.0256963 eV. This is much smaller than the
band gaps for CdSe and ZnO which shows how the temperature must be very high in
order for an electron to be excited thermally to the conduction band. In order for an
electron in CdSe nanoparticles to be excited the temperature would need to be 20189 K
and for ZnO nanoparticles the temperature would need to be 39450 K. These are very
high temperatures and rarely occur on earth, making it very unlikely that an electron will
be thermally excited to the conduction band.
3. The effective mass of an electron is impacted by the environment in which the electron is
in, and this is why it is different in a solid versus in a vacuum. This is because the
effective mass is the mass that can be observed when an electron is in an environment
where it is responding to forces present, such as those in a semiconductor. The effective
19

mass of a hole is the effective mass of the positively charged region that is deficient of an
electron.
4. See data analysis.
5. See data analysis.
6. It was found that when the me* values was dependent on 1/R and epsilon was dependent
on 1/R a much better fit for the experimental data was obtained using equation 4. Both
epsilon and me* were multiplied by 1/R to obtain the fit plotted in figure 12.

Figure 3: Adjusted fit of CdSe crystals where me* and epsilon are multiplied by 1/R to change the slope of the theoretical model.

20