The following document is an example lab report for you to use in constructing your own
reports. Please examine the report requirements as specified in your lab manual and compare
them with the format and content of the following document.
A Color Shifting Equilibrium of Cobalt
by Ken Overway for CHEM XXX, 6/12/2008
Abstract
Transition metal complexes are typically colorful because the differences between the d-orbital energy
levels match the energy of visible radiation. This makes studying some of their chemical properties easy
using spectroscopic methods. In this experiment the equilibrium between two cobalt complexes was studied
as a function of temperature in order to determine thermodynamic properties of the reaction. The
quantitative results confirmed that the reaction was endothermic and increases entropy, as expected by
inspecting the balanced reaction. Several experimental parameters were varied and studied in order to
explain variations in the experimental results. This experiment can employ digital data collection and was
utilized as a General Chemistry II lab.
Introduction
Transition metal complexes are typically very colorful because the differences between the d-orbital
energy levels match the energy of visible radiation. Cobalt(II) is a d7 transition metal that changes color
when involved in the following reaction
Co(H2O)62+(aq) + 4 Cl-(aq)  CoCl42-(aq) + 6 H2O(l)
(1)
where the octahedral aqua complex appears pink and the tetrahedral chloride complex appears blue. Crystal
Field Theory explains the color change as a result of two effects that change the orbital splitting energy.
The largest effect is the conversion from the octahedral to the tetrahedral complex, attenuating the splitting
energy to 4/9 of its original value. A secondary consideration is the displacement of a relatively weak-field
ligand (H2O) with a very weak-field ligand (Cl-) (1). The color observed is the complementary color of the
absorbed color, so if [Co(H2O)6]2+ has a larger splitting energy, its absorbed wavelength will be bluer and its
observed wavelength will be redder compared [CoCl4]2-.
The reaction under investigation is also temperature
complex and removing heat shifts it to the pink complex
Absorbance
dependent. Adding heat shifts the reaction to the blue
color absorbed
2.5
color observed
2
1.5
1
0.5
(Figure 1), making it possible to measure some
0
thermodynamic parameters of this reaction, namely entropy
and enthalpy. The equilibrium constant is related to the
2
400
500
600
Wavelength / nm
700
800
T=68.4 ºC
T=65.3 ºC
T=59.8 ºC
T=46.1 ºC
T=35.3 ºC
T=30.1 ºC
T=24.8 ºC
T=20.2 ºC
T=17.1 ºC
T=8.4 ºC
Figure 1. Spectral scans of the hexaaqua (520 nm)
and tetrachloro (600-700 nm) complexes of cobalt.
Gibb’s Free Energy of the reaction via
-Grxn = RT ln Keq
(2)
Since G is very temperature dependent, it can be replaced by enthalpy and entropy
G = H - TS
(3)
Substituting equations 2 into 1 and dividing both sides by RT yields a linear equation.
ln K eq 
y 
 ΔH 1 ΔS
 
R T
R
(4)
m  x b
By measuring the Keq as a function of temperature, equation 4 yields the H and S from the slope and yintercept.
Others who have studied this reaction have only focused on it as an example of Le Chatelier’s
Principle in a purely qualitative manner (2-4). These studies have students setup a series of test tubes, to
which they add various reagents that directly affect the concentration of the reaction species and observe
which way the reaction shifts. From these qualitative observations students can draw make conclusions
about whether the reaction is exothermic, antropic, etc. While these experiments are interesting, they do not
involve the use of complicate data collection equipment or require much data processing – both of which are
highly valued skills that our laboratory program seeks to develop. The purpose of this experiment is to
quantitatively determine the entropy and enthalpy of the cobalt reaction by studying the equilibrium as a
function of temperature.
Experimental
Chemicals needed for this experiment include cobalt chloride hexahydrate and concentrated
hydrochloric acid. A single spectrophotometer is required to measure the absorbance a cobalt solution.
Volumetric glassware, a hotplate, and appropriate 1 cm glass or plastic cuvettes were used to prepare, heat,
and measure the solutions.
Two different spectrophotometers where employed in this study for purposes of comparison. One
experimental setup was composed of a Milton Roy Spectronic 20D+ for absorbance measurements and a
Tenma 72-6870 Data Logger for the temperature measurements. A thermocouple (Fluke general purpose
bead probe 80PK-1) was modified with banana plugs for compatibility with the Data Logger. The wire-like
thermocouple was needed in order to conveniently monitor the temperature of the cobalt solution while the
3
cuvette was within the spectrophotometer. If digital data acquisition is desired, a data cable for the
Spectronic 20D+ and the data cable that is provided with the Tenma Data Logger need to be interfaced with
a computer. The authors used customized software constructed with Visual Basic to accomplish this. One
serial channel on the computer collected absorbance data from the Spectronic 20D+ while another serial
channel collected the temperature. The Vernier setup, described below, was found to be much simpler and
easier to use as the software and interfacing needs are provided.
The second spectrophotometer used was a Vernier Colorimeter, a thermocouple, and a LabPro unit.
The Logger Pro 3.0 software provided with the LabPro unit was able to collect both absorbance and
temperature using a single interface.
Part I. The Calibration Curve
A calibration curve of standards from 0.1 mM to 3 mM Co was prepared from a 15.33 mM stock
solution of Co in a concentrated HCl solution (0.1050 g of CoCl2•6 H2O in a 25mL volumetric flask) in
order to facilitate the determination of the concentration of the tetrachloro complex during the temperature
trials. It is assumed that the approximately 12 M HCl is sufficient to shift the equilibrium of reaction 1 to
the tetrachloro complex. While this assumption is not perfect, the validity of this assumption is irrelevant
since there are no literature values to make proper comparisons. These solutions were all prepared in the
hood using volumetric glassware. Absorbance measurements at 653 nm (one of the available wavelengths
of the Vernier Colorimeter) were made at room temperature using one of the spectrophotometers. A linear
fit of the absorbance vs. concentration data was used to determine the concentration of the tetrachloro
complex in the temperature trials.
Part II. The Temperature Trials
A 12.0 mM Co solution was prepared in 6 M HCl (0.0802 g of CoCl2•6 H2O in 25 mL volumetric
flask). The 6 M HCl provided the necessary balance between the tetrachloro and hexaaqua complexes of
cobalt in order to observe a dramatic color shift from 0 °C to 100 °C. Standardization of the 6 M HCl is
discussed later. Temperature trials involved heating the cobalt solution to 80 or 90 °C by placing a cuvette
containing the cobalt solution into a water bath. The temperature was monitored using the thermocouple. A
razor blade was used to cut a square hole in the cuvette cap of the plastic cuvette in order to insert the
thermocouple wire. The wire was bent such that it did not block the light path through the middle of the
4
cuvette. When the solution reached 80 ºC or more it was wiped down with lens paper and placed into the
spectrophotometer, where absorbance was also measured at 653 nm along with the temperature as the
solution cooled to room temperature. The cooling from 70 ºC to near room temperature takes about 20
minutes, so data collection could be accomplished without a computer if the absorbance and temperature
were written down at certain intervals without regard to consistent sampling rate.
Overheating the plastic disposable cuvettes in the hot water bath lead to warping of the cuvette,
which made it impossible to insert it into the Vernier colorimeter sample chamber. It was found that if
temperatures are kept below 80 ºC this problem is avoided. Glass cuvettes used with the Spectronic 20D+
do not suffer this problem and can be heated to higher temperatures.
The raw data files generated were named “calibcurveComp?.txt” and a series of filed named
“TTcomp2BhotNS0611A.txt” (see lab notebook for the cipher rules for these files). The entire dataset was
processed in “CobaltKeqDataSetJune2007.v2.xls”.
Results
1.400
High quality calibration curves can be obtained using
0.99 and slopes near 400 M-1cm-1, depending on the sensitivity of
the instrument (as seen in Figure 2). The tetrachloro complex
1.200
Absorbance
either of the spectrophotometers, yielding R2 values of at least
0.800
0.600
0.400
0.200
concentration reaches its maximum at high temperatures and the
0.000
absorbance attenuates with decreasing temperature. Since the
total cobalt concentration is known in the temperature trials and
the concentration of the tetrachloro complex can be determined
y = 349.02x + 0.0086
R2 = 0.9992
1.000
0
0.001
0.002
0.003
0.004
Concentration of Co (M)
Figure 2. A typical calibration curve of Co in a
12 M HCl solution, using the Vernier
Colorimeter in this case.
using the calibration curve, the hexaaqua complex concentration
-8
is determined by the difference. Using the equilibrium constant
the chloride concentration of the temperature trial solvent, the
-9
ln Keq
expression for reaction 1, the cobalt complex concentrations, and
y = -5732.8x + 8.0787
R2 = 0.998
-10
value of the equilibrium constant can be determined as a function
of temperature. When this data is plotted as ln Keq vs 1/T
according the equation 4, Figure 3 results. The slope yields the
enthalpy of the reaction and the y-intercept yields the entropy of
the reaction.
5
-11
0.0029
0.003
0.0031
0.0032
0.0033
1/Tem perature (1/K)
Figure 3. Plotting ln Keq vs. 1/T to determine
H and S, where the slope = -H/R and S/R
(where R = 8.3145 J/K mol), yielding in this case
H = 47,665 J/mol and S = 67.17 J/K mol.
This experiment was developed using both types of spectrophotometers mentioned and employed in
a second semester General Chemistry laboratory section using the Vernier equipment. Table 1 contains the
accumulated results comparing the experimentally varied factors.
Table 1. Comparison of results using 2-sample t-statistics
Condition
Spectronic 20D+
Vernier Colorimeter
Enthalpy /J mol-1
Std.
Avg.
Dev.
35,610 
380
34,700  3,500
Entropy /J K-1 mol-1
Std.
Avg.
Dev.
15.04 
0.84
14

11
N
2
2
tcalc
(H)
tcalc
(S)
tcrit @ 95%
conf.
0.37
0.17
4.30
CoCl2•6 H2O
Co(NO3)2•6 H2O
46,000
45,600


2500
2700
61.6
61.9


7.1
8.0
6
7
0.28
0.07
2.20
stirring – yes
stirring – no
46,000
45,700


1500
2800
62.3
61.6


4.3
8.1
3
10
0.17
0.13
2.20
5.66 M HCl
6.96 M HCl
79,100
45,400


6100
2400
157
61.1


18
6.6
9
16
19.79
19.14
2.07
cold to room temp.
62,000  11,000
115

37
4
5.94
5.97
2.10
hot to room temp.
45,400  2,400
61.1 ±
6.6
16
NOTE: N is the number of replicates, tcalc is the result of a 2-sample t-statistic, and tcrit is the critical value
Discussion
Individual trials (see Figure 3) show that the solution is extremely well behaved, producing a highly
precise measurement. When comparing consistency between trials, the relative standard deviation can still
remain under 5% when experimental conditions are identical. A number of conditions were considered
including spectrophotometer type, the stock cobalt compound (either cobalt nitrate or cobalt chloride),
stirring, HCl solvent concentration, and direction of temperature change.
Examining Table 1 reveals that both spectrometer types yield the same results within 95%
confidence. It is also not surprising that the source of the cobalt compound is irrelevant given the
insignificance of the concentration compared to the chloride in the solvent. Stirring had no effect on the
numerical results and shortened the cooling time only marginally while complicating the experimental setup
with the purchase of small stir bars and placing the spectrophotometer on top of a stir plate. It was also a
concern that the heat dumped into the spectrophotometer chamber may cause instrumental drift. Monitoring
the absorbance of a hot water cuvette in a room temperature spectrophotometer showed this was not a
significant factor.
6
As expected, the HCl concentration of the solvent in the temperature trials had a significant effect on
the results, both visually and mathematically. Since the term is raised to the 4th power in the equilibrium
constant expression it needs to be accurately determined. Two different concentrations were prepared for
the purpose of comparison. Each solvent was standardized using a sodium carbonate standard. The results
using the HCl solvent near 5.66 M HCl produce enthalpy and entropy values that were larger in magnitude
than the 6.96 M HCl solvent and was visually pinker, as expected from the Le Chatelier’s principle. This
same elevation of thermodynamic parameters occurred with the cold and hot temperature trials where the
initial cold solution was also pinker than the hot solution.
The final parameter examined was the direction of the temperature change. Heating the cobalt
solution and allowing it to cool to room temperature produced data sets that covered a larger temperature
range, produced more consistent results, and were preferred to the cold trials, which used an ice bath to cool
the cuvette and then allowed it to warm to room temperature in the spectrophotometer. The cold trials were
more difficult to execute than the hot trials due to cuvette fogging. In order to prevent fogging, dry nitrogen
gas was blown into the sample chamber of the spectrophotometer. This was successful at preventing the
fogging, but the cold trials were abandoned when the experiment was conducted in the General Chemistry
lab since they would have required the purchase and preparation of several drying tubes and the use of the
nitrogen gas lines in the hoods. The extra experimental complications in the cold trials did not yield
commensurate quality data sets. HCl solvent concentrations should be increased for cold trials if they are
used since the absorbance of the cold 6.96 M HCl solutions begin below the lowest standards of the
calibration curve and below the normally acceptable absorbance range of the spectrophotometers used due
to detector noise (5).
The chemical system under study in this experiment involves high ionic strength values and
therefore does not lend itself to comparison to literature values for the thermodynamic quantities at standard
state. One could attempt to determine the standard state enthalpies and entropies of formation for the two
cobalt complexes, but the calculations necessary would be beyond the intent of this laboratory experiment
designed for an introductory chemistry laboratory.
7
Conclusions
This experiment provides students with an opportunity to measure thermodynamic parameters of a
visually appealing reaction. It includes building the skills involved in solution preparation, calibration
curves, and extensive data processing. Instructors can determine whether students collect data visually or
digitally, which then also adds a layer of learning to use computer interfaced equipment and data
importation. The chemical requirements are minimal and the standardization of the temperature trial solvent
is unnecessary if a common HCl concentration is prepared for the entire laboratory section, which would
allow grading based on class averaged values for the enthalpy and entropy.
References
1. Miessler, G.L.; Tarr, D.A. Inorganic Chemistry, 3rd ed.; Prentice Hall: Upper Saddle River, 2004, pp
360, 368.
2. Grant, A.W. J. Chem. Educ. 1984, 61, 446.
3. Ophardt, C.E. J. Chem. Educ. 1980, 57, 453.
4. Barrera, N.M.; McCarty,J.L.; Dragojlovic, V. Chem. Educator 2002, 7, 142-145.
5. Skoog, D. A.; Holler, J. F., Nieman, T. A. Principles of Instrumental Analysis, 5th ed.; Harcourt:
Philadelphia, 1998, pp 307 – 311.
8