Brent Amberger '08 and Nazir Savji '08 ◗

Thermochromism of Transition Metal Compounds


Thermochromic behavior can occur through a number of different mechanisms. Here four compounds are examined. Bis(diethylammonium) tetrachlorocuprate (II) changes color as a result of a geometry change from square planar to distorted tetrahedral. bis(diethylammonium) tetrachloronickelate (II) changes color as a result of a decrease in coordination number from six (octahedral) to four (tetrahedral). Ag2HgI4 and Cu2HgI4 change color as a result of a change in lattice formation from a highly organized tetragonal to a more random tetragonal lattice. The mechanisms proposed by the literature are reinforced by magnetic susceptibility readings, UV-vis, mid IR, and eyeball spectroscopy.


A compound that displays a reversible change in color as a result of a change in temperature is a thermochromic compound. Many metal complexes as well as organic compounds are known to exhibit thermochromic properties both in the solid phase and when dissolved in water. These experiments focus on solid metal complex thermochromism.

Thermochromic behavior arises from many types of rearrangement at the molecular level. Various mechanisms include: phase transition, change in ligand geometry, equilibria between different molecular structure, and change in the number of solvent molecules in the coordination sphere.(1) In this report, changes in ligand geometry, coordination number, and lattice structure are shown to lead to changes in color. Thermochromic behavior can be divided into two main classes: continuous and discontinuous. Some color changes are gradual; as the temperature increases over time, the color slowly changes. This can be due to the breaking or rearrangement of the lattice. This type of rearrangement is also known as continuous thermochromism.(3) On the other hand, a dramatic change in color can occur at a specific temperature or over a very small temperature range. This is known as discontinuous thermochromism.(4)

In this report, changes in ligand geometry, coordination number, and lattice structure are shown to lead to changes in color.

The practical implications of thermochromic compounds are numerous. Liquid crystals change geometry gradually depending on the temperature. Therefore, they can begin black and exhibit all colors of the spectrum depending on the temperature.(5) They have applications in thermometers for rooms, kitchen appliances, and even medical uses. Although liquid crystals change geometry slowly, they are still classified as discontinuous thermochromic compounds because at different stages of the geometry change, the observed colors are very different.

Leucodyes operate quite differently from liquid crystals. They are much less accurate than other thermochromic compounds and are used where accuracy is not an issue.(6) An interesting example is battery state indicators.(7) A layer of leucodye is applied on top of a strip and then each end is attached to the ends of the battery. Since the strip has a resistance, it heats up when the current of the battery travels through it, which is displayed by a change in color of the leucodye. Since the strip is triangular, its resistance is changing as you move along the triangle, and thus is heats up more at certain points along the triangle than at others. Therefore, the higher temperature parts of the triangle change the color of the leucodye and tell the consumer how much life is left in their battery.

Thermochromic clothing like that developed by Hypercolor and toys that change color from the heat of a child's hands are fun and interesting, but thermochromics can also have more important applications. They have become important in trying to respond to changes in the environment. Thermochromic compounds could be used in a window system to control the flux of solar energy to control heating by the sun. Another important application of thermochromic compounds that has been suggested is road signs that change color when the temperature dips below freezing, alerting motorists of possible icy conditions.(8)

Compound Analysis: bis(diethylammonium) tetrachlorocuprate (II)

This compound is a bright green color at room temperature and undergoes a reversible solid-state reaction to a yellow color at temperatures above 52?C-53?C.(9) At temperatures below 52?C-53?C, the four coordinate compound has a square planar geometry even though it is not a d8 compound. Because of the short hydrogen bonds between the hydrogens of the counter ion, diethylammonium, and the chlorines on the metal compound, it favors this square planar geometry. In the planar CuCl42- compound, the NH---Cl bond distance is 331 pm. When the temperature passes 53?C, the color changes to yellow because the hydrogen bonds lengthen to approximately 345 pm and the geometry changes to the much less constrained distorted tetrahedral.(10) The change in geometry changes the arrangement of the d orbitals as seen in Figure 1.

Figure 1: The change in geometry changes the relative energies of the d electron orbitals. Square planar arrangement is pictured on the left and tetrahedral on the right.

There is no stable intermediate between the square planar and the distorted tetrahedral geometry, so the change is an example of discontinuous thermochromism. Individual clumps of crystals appear as either green or yellow depending on their exact temperature.

The synthesis of this compound was done using diethylammonium hydrochloride and copper (II) chloride as outlined by Michlel J. M. Van Oort.(11) The resulting compound formed green crystals, which were filtered out of the remaining isopropyl alcohol and dried in a desiccator for two weeks. (The literature only suggested drying them overnight before analysis.) The compound was stored in a desiccator with several other not-yet dry compounds, and during this time some of the crystals discolored to a darker/browner green. These crystals were presumed to be contaminated, and they were physically separated from the rest with a spatula. It was suspected that the compound might have still been somewhat contaminated though because it was a somewhat darker color than it should have been, and had a pasty consistency instead of being powdery.

When heated, the compound changed quickly from green to yellow around 50-54 C, with individual crystals either being green or yellow, with no green-yellow in between. See Appendix A.

Figure 2: UV/vis for [(CH3CH2)2NH2]2CuCl4

Figure 2 shows the visible absorption spectrum for [(CH3CH2)2NH2]2CuCl4 The UV/vis was roughly consistent because the lowest absorbance is at approximately 565 nm } 5 nm, which is the wavelength of green light. Since our major peaks are in the red and violet region of the spectrum, the compound is expected to be green-yellow in color, when in fact it is solely green. The peak in the red region of the spectrum was also expected to be larger than the peak in the violet region of the spectrum.

The infrared (IR) spectrum for the room temperature tetrachlorocuprate can be seen in Appendix C. It seems that there are two small nearby peaks in the 3100 cm-1 region, which is consistent with the literature.(12) Had an IR been run for the heated compound, it would be expected to display less splitting in the same region because the weakening strength of the NH---Cl hydrogen bond and hence the strengthening of the N-H bond.

A magnetic susceptibility test at low temperature resulted in an n value of ~ .35. The compound should be paramagnetic with an n value of 1. Given the paste-like consistency of the compound and the difficulty in properly loading the test tube, the result of the test is relatively consistent with the expected value.

Compound Analysis: bis(diethylammonium) tetrachloronickelate (II)

At low temperatures this compound is a light yellow-brown color, and when heated to temperatures over 72-73 C it changes to a brilliant bluish color. In the low temperature arrangement, neighboring molecules share bridging chlorines. The molecules are arranged in a two-dimensional sheet such that each compound shares two of its chlorines with nearby molecules, and has two chlorines shared with it. In effect, the low temperature form of this compound has an octahedral geometry. As temperature increases, higher molecular motion breaks the bridging bonds, and each transition molecule compound becomes four coordinate.(13) NiCl42- is a d8 high spin complex, and therefore favors tetrahedral arrangement. The breaking of the 2-d sheet is an example of continuous thermochromism. With increasing temperature the sheets become smaller and smaller with more and more lone four coordinate tetrahedral ions.

Figure 3: As temperature changes, the coordination number and therefore geometry of each Ni ion changes from a six coordinate octahedral to a four coordinate tetrahedral

The change from octahedral to tetrahedral changes the configuration of the d orbitals as seen in Figure 4. Because the low temperature is yellow and the high temperature is blue, the energy of the electronic transition in the low temperature state must be larger (roughly double) than in the high-energy state. This is certainly consistent with the fact that deltao is larger (roughly double) than deltat.

Figure 4: As the coordination number changes and the complex goes from octahedral (left) to tetrahedral (right), the relative energies of transition decreases.

The compound was synthesized according to instructions in the article by Changyun et al. After the crystals had been heated in the oven for an hour they were stored in a desiccator for a week before they were analyzed.

As expected, the compound showed a color change around 73-76 C. Also as expected, the compound was continuous thermochromic, with the color changing from light yellow to brighter greens and blues.

The UV-vis spectrum of the compound at low temperature shows a peak in the blue/violet region. This is logical since the compound is yellow.

Figure 5: UV- vis spectrum for low temperature bis(diethylammonium) tetrachloronickelate(II)

The magnetic susceptibility for the compound in the low temperature state yielded an n value of ~.65. It should have been 2, so although the paramagnetic trend is correct, the prediction of one unpaired electron instead of two is strange.

Compound Analysis: Ag2HgI4 and Cu2HgI4 (M2HgI4)

The structure and behavior of these two compounds is very similar. Ag2HgI4 is a yellow color at low temperatures and changes to an orange color at temperatures over 50 C. Cu2HgI4 is a red color at low temperatures and changes to a purple color at temperatures over 67 C. Both colors are very strong. HgI42- is a d10 compound, which has a tetrahedral shape. It forms a lattice with its metal (I) counter ion. The crystal lattice can be viewed as face centered tetragonal. The HgI42- ion occupies all 8 corners, while the M+ occupies 4 of the 6 faces, with the other two faces empty (see Figure 6 below). At higher temperatures, the lattice is re-organized such that some of the M(I) and Hg(II) ions switch places (though iodide ions stay in the same positions). The color apparently changes as the lattice goes from highly ordered to a more random distribution, which is favorable because entropy is increasing. This change from order to disorder is continuous and hence the thermochromism is too. There are intermediates with somewhat random distribution of ions in the lattice, which have a middle yellow-orange or red-purple color. The electronic source of the brilliant color is not known exactly, but the fact that all the metals involved are d10 rules out d-d transitions. An electron transfer of some kind might be occurring, especially since the colors observed are so intense. Chloride is a terrible pi acceptor, so the color is not metal to ligand charge transfer. The color could be ligand to metal or metal to metal charge transfer, although the electron would have to enter something other than the d orbitals (possibly an s orbital). As the temperature increases and the lattice is rearranged, the M(I) and Hg(II) ions remain in the same environment relative to the iodide ions, but their environment in relation to the other metal ions is changing. As such, a metal-metal charge transfer seems the most likely. Also, because Ag2HgI4 and Cu2HgI4 have different colors, the charge transfer is probably between the M(I) and the Hg(II).

Figure 6: Low temperature arrangement of tetragonal crystal lattice. At high temperatures some Hg(II) and M(I) ions interchange.

Synthesis was done according to instructions given in Jeffrey Hughes' article.(14) After the red and yellow crystals were filtered and collected, they were stored in a desiccator for a week before analysis.

As expected, Ag2HgI4 changed color around 47-51 C. The color change was gradual, with intermediate yellow-orange colors. The Cu2HgI4 compound changed color around 68 -69 C, with intermediate red-purple compounds. Appendix A shows these transformations, although the difference in color was more visible than what is seen in the pictures.

Figure 7: Even when diluted in clear salt, these compounds gave absorptions that were off the scale.

The UV-vis data from these compounds show absorbances that are way off the scale, and therefore the position of the peaks cannot be determined, but at least it verifies that the compounds have very intense colors (as a result of these high absorptions). Not surprisingly, magnetic susceptibility tests showed both of these compounds to be diamagnetic, as a result of every metal being d10


The analysis of these compounds shows that changing temperature reliably lead to a change in color for certain compounds. In these experiments, heat lead to three very different types of atomic rearrangement, which in turn changed the energies of electronic transitions.

Several aspects of our experiment could be changed for improvement in the future. All compounds should be stored in separate desiccators. This will prevent gases from altering the properties of our solids. Furthermore, solid-state UV-vis spectrometer holders must be better tested to improve the quality of spectral data. Also, IR and UV-vis spectra should have been obtained for not only the room temperature compound, but also the heated compound while it had changed color. This would have provided data for proper analysis, which could have been compared against the literature.


We gratefully acknowledge the kind help of Professor Burkett and Khadine Higgins for their help in making sure everything we needed was available to us.

References ◗

1. Leblanc, M.; White, M. A. J. Chem. Educ. 1999, 76, 1201.

2. Bloomquist, D. R.; Pressprich, M. R.; Willet, R. D. J. Am. Chem. Soc. 1988, 110, 7391.

3. Bloomquist, D. R.; Dodgen, H. W.; Roberts, S. A.; Willett, R. D. J. Am. Chem. Soc. , 1981, 103, 2603.

4. Van Oort, M. J. M. J. Chem. Educ. , 1988, 65, 84.

5. Leblanc, M.; White, M. A. J. Chem. Educ. 1999, 76, 1201-2.

6. Leblanc, M.; White, M. A. J. Chem. Educ. 1999, 76, 1202.

7. Middlesex Univerisity. Smart Colours. .

8. Leblanc, M.; White, M. A. J. Chem. Educ. 1999, 76, 1204.

9. Haugen, J. A.; Leback, J.; Willett, R. D. Morrey J. Inorg. Chem. 1974, 13, 2510.

10. Choi, S.; Larrabee, J. A. J. Chem. Educ. 1989. 66. 775.

11. Van Oort, M. J. M. J. Chem. Educ. , 1988, 65, 84.

12. Choi, S.; Larrabee, J. A. J. Chem. Educ. 1989. 66. 775.

13. C. Changyun, Z. Zhihua, Z. Yiming, D. Jiangyan, J. Chem. Educ. , 2000, 77, 1206.

14. Hughes, J. G. J. Chem. Educ. , 1998, 75, 57.?

About the Essay ◗

This report was the culmination of 4 weeks of independent laboratory work where we examined the thermochromatic properties of four different transition metal compounds. After reading several papers relating to thermochromatic compounds, we used different chemical and spectroscopic tests to determine the characteristics of the compounds that make them thermochromatic. We also gave a 20-minute PowerPoint presentation to our classmates.

About the Authors ◗

Brent Amberger (left, a native of Maine) and Nazir Savji (right, a native of Ontario) are chemistry majors from the class of 2008. They have known each other since freshman year, where they both lived in Valentine and have taken more than six classes together. Brent is currently abroad in New Zealand and will be conducting his thesis with Professor Leung next semester. Nazir is a Student Health Educator and Resident Counselor at Amherst and will be conducting his thesis with Professor Marshall next semester.