Hemes are the iron complexes of aromatic tetrapyrrole macrocycles called porphyrins. Hemes and heme proteins are vital components of essentially every cell of almost every living organism. They have many diverse roles in vital life processes, including the transport and storage of oxygen in higher animals (hemoglobin and myoglobin, respectively); sensing of oxygen in nitrogen-fixing bacteria (FixL); regulatory functions based on nitric oxide (nitrophorins, guanylyl cyclase); electron transport in the respiratory chains of organisms as diverse as bacteria, yeasts, plants and animals, and in photosynthetic cells from those of the simplest photosynthetic bacteria to those of higher plants (cytochromes a, b, c, d, f, o); synthesis, modification and/or degradation of fatty acids, steroid and adrenal hormones, anesthetics, xenobiotics (cytochromes P450), nitric oxide (NO synthase)and carbon monoxide (heme oxygenase); activation and metabolism of hydrogen peroxide (peroxidases, myeloperoxidase, haloperoxidases, catalases, etc.); and metabolism of the oxides of nitrogen and sulfur (nitrite reductase, sulfite oxidase, etc.). Biologically important related molecules that involve modified hemes include the reduced hemes (iron chlorins, bacterio-, isobacterio- and dioxoisobacterio-chlorins, and sulfhemes), as well as chlorophylls (Mg²⁺ tetrapyrrole complexes) and pheophytins (metal-free reduced tetrapyrroles), vitamin B₁₂ (Co(I,II,III) tetrapyrrole complexes) Factor F430 (a Ni tetrapyrrole complex), and some of the luciferins responsible for the diurnal cycles of many species of plants and animals. Breakdown products of heme (the bile pigments) are harmful to newborn infants and alcoholics, and intermediates in the synthesis of heme are harmful to persons suffering from various forms of porphyria. Genetically modified forms of hemoglobin are responsible for sickle cell anemia and a number of methemoglobinemias. Synthetically prepared or naturally occurring hemes are utilized as contrast agents in medical imaging techniques (CAT, PET and MRI) and as photosensitizers to create singlet O₂ for photodynamic therapy techniques utilized to destroy tumors.

Among these many roles of hemes and heme proteins are the oxygen-binding, carrying and storage roles of hemoglobin (present in the blood streams of all higher animals) and myoglobin (the storage protein found in the muscles of all higher animals). The oxygen-binding equilibria, kinetics and spectroscopy of these and related heme proteins have been studied by many scientists for well over fourty years. It was learned quite early that the active, reversible oxygen-binding forms of heme proteins were five-coordinate Fe(II) heme centers that could bind O₂ to the remaining site:

Fe(II) has a d⁶ electron configuration, and in the presence of a square pyramidal ligation geometry it is paramagnetic, S = 2, four unpaired electrons, otherwise known as high-spin Fe(II). (In fact, Fe(II) porphyrins have a different spin state for each coordination number: S = 1 for 4- coordinate, S = 2 for 5-coordinate, and S = 0 for 6-coordinate. Can you explain why, using crystal field theory?) Molecular O₂ is also paramagnetic, S = 1, with two unpaired electrons in the p* orbitals. The product, the dioxygen complex of the heme center, in which the iron is six-coordinate and O₂ is unsymmetrically bound to Fe, is diamagnetic, although there is still an argument about whether the electrons are all paired or whether the electron configuration might be low-spin Fe(III) (d⁵, S = ½) bound and antiferromagnetically coupled to the superoxide radical, O₂`(S = ½) resulting from electron transfer from Fe(II) to O₂.

Much of the information available about the bonding and electron distribution in the dioxygen complexes of the heme centers of heme proteins has been obtained by studying synthetic hemes, in particular, the iron complexes of tetraphenylporphyrin anions, TPP²`. (The structure of the free base tetraphenylporphyrin, H₂TPP, is shown in Figure 1.)

However, the Fe(II) complexes of tetraphenylporphyrins (as well as the natural porphyrins when not sequestered by the proteins) are so reactive toward molecular oxygen that they rapidly react with it irreversibly:

In fact, it has been said many times that one of the important roles of the proteins is to keep the hemes well separated, and thereby prevent two iron(II) porphyrins from reacting with each other in the presence of molecular oxygen, first to form the µ-peroxo dimer, which eventually decays to the µ-oxo dimer.

One solution to the extreme reactivity of Fe(II) porphyrins with molecular oxygen, which has been used extensively to understand the first (reversible) step in O₂ binding, is to use Co(II) as a substitute for Fe(II). When either 4-, 5- or 6-coordinate, Co(II) porphyrins are low-spin d⁷ centers (S = ½) that, like their Fe(II) counterparts, when present in the active site of a heme protein, react reversibly with molecular oxygen. However, the single unpaired electron of Co(II) can only pair with one of the two unpaired electrons of molecular oxygen, so that the Co-O₂ adduct is of necessity paramagnetic (S = ½), and thus EPR active. Therefore, EPR spectroscopy is a very useful tool for studying Co(II) complexes in general, and especially Co-O₂ centers. Because the unpaired electron of low-spin d⁷ Co(II) porphyrins is in the d_z2 orbital, the EPR spectra are very sensitive to the binding of all sorts of ligands to the axial positions of the Co(II) porphyrin.¹ Co(II) porphyrins can readily be substituted for Fe(II) porphyrins in a wide range of heme proteins,² where they bind reversibly to molecular oxygen, although their complexes are not quite as stable as those of their Fe(II) counterparts. Cobalt(II) porphyrins are also excellent models for the " free radical" forms of B₁₂ enzymes as well, and a reversible oxygen complex has also been observed for this important biomolecule. ³ In the case of Co(II) porphyrins, the good news, as compared to Fe(II) porphyrins, is that in the absence of the proteins, the Co-O₂ complexes do NOT react irreversibly to form Co(III) µ-oxo dimers or other Co(III) products when in solutions containing molecular oxygen, at least on a time scale of hours, and in fact, in order for oxidation to occur, additional components must be present (H⁺, H₂O). We may explore this oxidative reactivity later in an advanced-level experiment (Section 6). For the present, we will prepare a Co(II) tetraphenylporphyrin and characterize it by optical, mass and ¹H NMR spectroscopy (yes, even though this is a paramagnetic complex, we can readily obtain its ¹H NMR spectrum ⁴), and then investigate its EPR spectra in the absence and presence of potential axial ligands and molecular oxygen. Because cobalt is a pure-isotope element (⁵⁹Co, 100%, I = 7/2), we will be able to use the size of the ⁵⁹Co hyperfine splittings, A_||Co, to estimate the relative spin density of the odd electron at the ⁵⁹Co nucleus for the 5-coordinate Lewis base complex and its O₂ adduct, and thus whether the latter is best formulated as Co(II)-O₂ or Co(III)-O₂ `.

Although this or a closely-related synthetic tetraphenylporphyrin free base will be supplied, the synthesis is included here in the event that a student wishes to prepare a differently-substituted tetraphenylporphyrin.

Place a Teflon-coated stir bar in a two-necked 250 mL round bottom flask fitted with a heating mantle, and add 40 mL of propionic acid (STENCH)! Fit the vertical neck of the flask with a reflux condenser using a TINY amount of grease on the joint. Be sure to close the other neck with a glass stopper. Heat the propionic acid almost to reflux and then add, sequentially, 1.57 mL of freshly distilled benzaldehyde (15 mmol) and 1.0 mL of freshly distilled pyrrole (15 mmol) using a syringe or automatic pipet. ( Caution! The solution is hot! Handle carefully!) Replace the glass stopper and heat the propionic acid to reflux. Continue to reflux the solution for 30 min, and then turn off the heating mantle and magnetic stirrer and allow the mantle and the solution to cool to room temperature slowly. (In the best of situations, leave the flask alone until the next laboratory period, for crystals of the H₂TPP form very slowly.) Filter the brownish purple mixture using a sintered glass frit. Wash the crude product with methanol to remove the tarry impurity. Rinse the mixture until the filtrate is clear and purple crystals are left on the filter. It is very important that all of the impurity is removed at this stage. Discard the wash solution and allow the crystals to dry by pulling air through them. Weigh the product and record the yield. NOTE: Free-base porphyrins are light-sensitive, and should be stored in the dark.

Determine the purity of the product (or the sample supplied) by spotting a s mall amount of a dichloromethane solution of your product on a silica gel TLC plate. To do this, dissolve a few crystals in about 0.5 mL of dichloromethane in a small vial, making sure that the sample is really dissolved. Use a microcapillary tube for spotting the solution on the plate. Place the TLC plate in a screw-capped jar. Elute with a 1:1 mixture of toluene and hexane. (Before inserting the TLC plate, make certain that the eluting solvent level is below the spot on the plate.) If the developed TLC plate indicates that you have significant (colored) impurities, then the H₂TPP can be recrystallized by dissolving the solid into 100 mL of CH₂Cl₂, filtering the solution, then adding 40 mL of methanol to the filtrate and concentrating the solution to 50 mL on a rotary evaporator, then placing (stoppered) in the refrigerator overnight. Suction-filter the resulting crystals. Record the ¹H NMR spectrum of your product in CDCl₃ from -5 to +15 ppm. Then add 2 drops of D₂O, stopper and shake the NMR tube vigorously for about 15 seconds, and again record the NMR spectrum. (Is there any change in the spectrum? Explain.) Record the UV-vis spectrum using methylene chloride as a solvent. NOTE: Absorbance should be between 0.1 and 1.2 for the peaks of interest to you. Therefore, you will need to obtain a spectrum of the 470-700 nm region as a more concentrated solution, then dilute it by a factor of about 20, and re-record the spectrum, so that the intense "Soret" peak at about 420 nm, as well as the peaks in the UV region, will be on scale. Finally, submit a sample to obtain a mass spectrum of your product.

Using your pure sample of H₂TPP, place 0.2 g (0.33 mmol) in a 250 mL 2-necked round bottom flask equipped with a Teflon stir bar, a condenser, a stopper for the other neck, and a heating mantle. Add 80 mL of fresh (newly opened bottle or freshly distilled) chloroform and stir the mixture to dissolve the H₂TPP, and begin heating. While this is taking place, make a solution of 1.0 g of cobalt acetate tetrahydrate (4.0 mmol) in 50 mL methanol and heat to near boiling, until the Co(II) acetate is dissolved. Remove the stopper from the round bottom flask and pour in the methanol solution of Co(OAc)₂(H₂O)₄, replace the stopper and heat to reflux. Continue refluxing and stirring for 30-60 minutes, occasionally checking for the disappearance of strong red fluorescence under long-wave UV light (CoTPP does not fluoresce, while the free-base does), or remove tiny samples of the solution, spot on TLC, allow to stand until the methanol has evaporated, and then elute with toluene/hexane as above. The CoTPP should move ahead of the H₂TPP, and CoTPP does not fluoresce. Insertion of Co(II) is complete when no spot of the free base porphyrin can be detected. Additionally, UV-visible spectroscopy can be used to determine the completeness of metal insertion, for although the free-base porphyrin has four bands between 470 and 700 nm, the Co(II) porphyrin will have only two, both of which are at shorter wavelength than the longest-wavelength band of the free-base porphyrin.

When metal insertion is complete, turn off the heat, remove the heating mantle, and allow the solution to cool to room temperature. Place about 100 mL of distilled water in a separatory funnel, and pour the cooled reaction mixture in on top of it. DO NOT SHAKE!!! The chloroform will sink to the bottom, while the methanol and excess inorganic salt will dissolve in the water. WITHOUT SHAKING, draw off the chloroform layer, discard the aqueous layer, place a fresh 100 mL distilled water into the separatory funnel, and again pour the reaction mixture into the funnel, allowing the chloroform layer to collect at the bottom of the funnel. Again, draw off the chloroform layer without shaking. At this point, MOST of the inorganic salts, and most of the methanol, have been removed, and the chloroform solution of the cobalt porphyrin can be washed several additional times (now with shaking of the separatory funnel) with fresh water. Dry the chloroform solution over Na₂SO₄, filter off the drying agent, and evaporate to dryness, transfer to a weighed vial, and obtain the weight and measure the yield of the product. The purity of the CoTPP can be determined using TLC, as described in section 2. If the starting H₂TPP was recrystallized before use and if you fully inserted cobalt, the product should not have to be further purified. However, if additional, different-colored band(s) is(are) observed in the developed TLC, the CoTPP will need to be purified by column chromatography.

Using TLC, determine which solvent system will allow you to separate CoTPP from H₂TPP and other impurities by column chromatography. With a very small portion of the CoTPP mixture, prepare a concentrated dichloromethane solution to use in the TLC trial separations. Since the best separations are achieved when this spot is the smallest, the dichloromethane solution should be applied with a very small capillary. This can be prepared by heating a disposable pipette over a low flame and quickly pulling the ends apart without sealing the heated part. Scratch the capillary with a sharp file and break the tube into two applicators. Dip an applicator into the dichloromethane solution of the mixture and touch it to the TLC plate, giving a spot that is no larger than 3 mm in diameter. Allow the dichloromethane to evaporate, and then make a second application of the solution to the same spot to increase the amount of CoTPP and impurities on this spot. The CoTPP and impurities are all intensely colored and are easily visible even at low concentrations on the plate. Make a drawing in your research notebook of the location of the spots for each of the five attempts. To establish which spot, after development, is H₂TPP, prepare another TLC slide, spotting it first with the dichloromethane solution of the mixture and then, in an adjacent but not overlapping position with a dichloromethane solution of pure H₂TPP. Develop the slide in one of the solvents that gave a good separation of spots, and establish which spot of the mixture is H₂TPP by comparing it with the known H₂TPP spot. From the TLC experiments you should select a solvent for the column chromatographic separation. Note the R_f values for different solvents in your lab notebook. You may select a solvent in which one of the components moves rapidly and the other more slowly. Such a solvent should give a good separation on the column. Alternatively, one might choose an initial solvent in which only one component moves while the other remains at the starting point. If the component which moves is your desired CoTPP, then it can simply be eluted and the impurities left on the column. Clearly, your strategy must be consistent with your observation as to whether your desired product moves fastest or more slowly than the impurities. A good thing to keep in mind is that all substances move more rapidly on a large column than they do on a TLC plate, so you should strive to find a solvent that will maximize the separation of your desired product from the other compounds present. Clearly, if you find that there is no H₂TPP and little or no other impurity in your CoTPP sample, then there is no reason to do column chromatography!

Clamp a chromatography column (3X30 cm) to a ring stand or other firm support. Push a marble-sized wad of glass wool to the bottom of the tube with a glass rod. Then pour enough washed sand to give a 1 cm layer. In a medium sized beaker, prepare a slurry of silica gel in your selected solvent. Be sure to "wet" the silica gel well enough with solvent, and stir with a glass rod, to insure that no air bubbles remain in it. Use enough silica to fill the column to a height of about 9-12 cm. Pour the slurry gently into the column and let the excess solvent drain until the meniscus is about 0.5 cm above the settled silica gel. (Do not allow the solvent level to drop below the top of the silica.) Carefully pour a slurry of silica and the CoTPP mixture in a few mililiters of the initial solvent onto the column. (The CoTPP should be soluble in this solvent; otherwise, a poor separation will result, as the remaining undissolved CoTPP slowly goes into solution, which will spread out your band.) Lower the solvent level again to the top of the silica gel and add another 1 cm layer of washed sand to prevent the bed from being disturbed when the eluting solvent is added. Then gently fill the column with the initial eluting solvent, being careful not to agitate the sand and bed beneath it. Carry out the elution, using a flow rate of about 1-2 drops per second. Insoluble residues, if any, will remain on top of the column; a sharp band or bands will remain near the top and move slowly down the column. This band is composed of impurities, principally H₂TPP, traces of tar, and any oxidized (Co^IIITPP) complexes having unknown axial ligands. The desired Co^IITPP product moves rapidly down the column as an orange smear and can be collected continuously until all the orange material has been eluted, or until the sharp band is about 2 cm from the bottom of the column. Collect only the orange band due to CoTPP. Concentrate the CoTPP solution to 25 mL and add 25 mL of methanol to precipitate the product. Filter the solution, rinse the product with a small amount of methanol, and dry it under vacuum. Record the weight and calculate the yield. The tar and any unknown Co^IIITPP complexes that remain at the top of the column should be discarded along with the used silica gel. HOWEVER, before dumping the used silica gel into the waste bag, allow the solvent to evaporate in the hood until the used silica gel is very dry, in order to avoid solvent pollution in the waste chemicals room. Characterize the CoTPP as described in Section 5 below.

As for the free-base, H₂TPP, the CoTPP should be characterized by UV-vis, ¹H NMR and low-temperature EPR spectroscopy, and by mass spectrometery. The appropriate solvent for all of these is toluene (d₈-toluene for NMR). The IR spectrum can also be obtained, as a mull or KBr pellet, but as you will see, there are many, many IR bands for both the free-base and metal porphyrin, which you cannot assign without making it a very major project. And because the O₂ complex is not stable at room temperature, no characteristic O-O stretch can be observed. If you wish to make the NO adduct, then the IR spectrum should show a highly characteristic N-O stretch in the region of 1500-1600 cm^-1.¹¹

For UV-vis, ¹H NMR and MS measurements, all of which are carried out at room temperature, you will not have to worry about the reversible reaction of CoTPP with molecular O₂. HOWEVER, for EPR studies, which are carried out at 77 K, O₂ dissolved in the solvent (toluene) will readily bind to the metal as the solution is cooled. Therefore, to obtain the EPR spectrum of CoTPP in toluene and toluene plus the desired Lewis base (a pyridine, imidazole, O-donor such as THF, DMF or DMSO, or a phosphine such as trimethylphosphine) NOT bound also to O₂, you must scrupulously remove the O₂ from the solution! Prepare one set of samples that are NOT de-gassed and another that have been very carefully de-gassed. You may want to use the glove box to accomplish this. EPR tubes are checked out from the stockroom (OC 106), and can be returned after use, assuming that you clean and dry them well. The ⁵⁹Co hyperfine coupling constant, A_||Co, is a measure of the amount of unpaired electron density at the nucleus of the cobalt. From the value of A_||Co for each of the complexes in the ABSENCE of O₂ as compared to the value in the PRESENCE of O₂, calculate the relative percent of the unpaired electron that is seen by the cobalt nucleus in the O₂ adduct, assuming that the hyperfine coupling constant in the absence of O₂ is a measure of the value when the electron is located 100% on the cobalt. If the percentage is less than 100%, where do you think the unpaired electron is? How could you test your hypothesis?

Co(III) porphyrins are octahedral low-spin d⁶ (diamagnetic) complexes that are excellent models for the Fe(III) centers of the important biological electron-transfer proteins known as the cytochromes. These electron-transfer proteins change the oxidation state of the metal reversibly from Fe(II) to Fe(III) and back again, with very little change in the Fe-N bond lengths of either the porphyrin nitrogens or the axial ligands. Typical axial ligands to the heme centers of the cytochromes are the imdiazole nitrogen of histidine, the thioether sulfur of methionine, and occasionally some other ligand, such as the phenolate oxygen of tyrosine or N-terminal amino nitrogen of the protein (cytochrome f of Photosystem II of higher plants). The cytochromes b of the mitochondrial Complex III and chloroplast cytochrome b₆f both have bis-histidine coordination, as does cytochrome a of mitochondrial Complex IV (cytochrome oxidase). On the other hand, mitochondrial cytochrome c, the only soluble, not membrane-bound, heme protein in the mitochondrial inner membranes, has one histidine and one methionine as axial ligands, as do many other cytochromes c, and a bacterial protein called bacterioferritin has bis-methionine coordination to its heme. It is very difficult to make synthetic mixed-axial ligand complexes, such as those that would be models of cytochrome c , without covalently attaching the ligands to the porphyrin. Furthermore, since one of our characterization methods will be ⁵⁹Co NMR, and since the ⁵⁹Co nucleus is quadrupolar and gives unobservably broad lines if its coordination sphere has symmetry lower than tetragonal (D_4h point symmetry), we will make only bis-L complexes of CoTPP for further characterization. Our goal in this part of the project is to prepare some bis-ligand complexes of Co(III) tetraphenylporphyrins, assign their ¹H and ¹³C NMR spectra, and record their ⁵⁹Co NMR spectra as a function of protonation/deprotonation of the N-H of the coordinated imidazoles (if used), in order to learn how each of these nuclei respond to H-bonding or deprotonation of the imidazoles. This is a true research project, and the results cannot readily be predicted.

Synthesis of the Co(III) porphyrin bis-ligand complexes begins with the Co(II) porphyrin products produced in the earlier parts of this project. See your laboratory instructor to obtain your product or another compound. The Co(III) porphyrins are produced in solution under reflux in the presence of the chosen ligands and AgBF₄ by air oxidation of the Co(II) starting materials. ¹² Follow the procedure given in reference 12; column chromatography is not necessary.

Characterize the complex(es) prepared in (deutero)chloroform by UV-visible and ¹H NMR spectroscopies. Discuss your results with your instructor before continuing with additional ¹H, ¹³C and ⁵⁹Co NMR spectroscopic investigations. Note that while some Co(III) porphyrins have reasonably sharp ⁵⁹Co resonances,¹³ others have extremely broad resonances in solution.¹⁴ Your products may be sent to other investigators for solid state NMR investigation.

1. "An ESR Study of Coordination to the Fifth and Sixth Position of "a,b,g,d-Tetra-(p- methoxyphenyl)porphinatocobalt(II)," Walker, F. A. J. Am. Chem. Soc. 1970, 92, 4235.

2. "Metal Substitution in Hemoglobin and Myoglobin," Hoffman, B. M. In The Porphyrins, Dolphin, D., Ed.; Academic Press: New York, 1979; Vol. VII, pp. 404-444.

3. "Superoxocobalamin, the First Step in the Autoxidation of Coenzyme B₁₂," Bayston, J. H.; King, N. K.; Looney, F. D. Winfield, M. E. J. Am. Chem. Soc. 1969 , 91, 2775.

4. "Proton NMR and ESR Investigation of Axial Solvation in Planar, Low-Spin Cobalt(II) Porphyrin Complexes," La Mar, G. N.; Walker, F. A. J. Am. Chem. Soc. 1973, 95, 1790.

5. Adler, A. D.; Longo, F. R.; Finarelli, J. D. J. Org. Chem. 1967, 32, 476.

6. Adler, A. D.; Longo, F. R.; Varadi, V. Inorg. Synth. 1976, 16, 213.

7. Badger, G. M.; Jones, R. A.; Laslett, R. L. Aust. J. Chem. 1964, 17, 1028.

8. "Synthesis, Spectroscopic and Structural Studies of Extremely Short Chain Basket Handle Porphyrins and Their Zinc(II) Complexes," Simonis, U.; Walker, F. A.; Lee, P. L.; Hanquet, B. J.; Meyerhoff, D. J.; Scheidt, W. R. J. Am. Chem. Soc. 1987, 109, 2659.

9. "Steric and Electronic Effects in the Coordination of Amines to a Cobalt(II) Porphyrin," Walker, F. A. J. Am. Chem. Soc. 1973, 95, 1150.

10. "Reactions of Monomeric Cobalt-Oxygen Complexes. I. Thermodynamics of Reaction of Molecular Oxygen with 5- and 6-Coordinate Amine Complexes of a Cobalt(II) Porphyrin," Walker, F. A. J. Am. Chem. Soc. 1973, 95, 1154, 7928.

11. Scheidt, W. R.; Hoard, J. L. J. Am. Chem. Soc. 1973, 95, 8281.

12. "Hydrogen Bonding to Coordinated Imidazole. Association of 1,10-Phenanthroline and Other Bases with Bis(imidazole) metalloporphyrins," Balch, A. L.; Watkins, J. J.; Doonan, D. J. Inorg. Chem. 1979, 18, 1228.

13. "Application of ⁵⁹Co NMR to Cobalt(III) Porphyrins. Linear Relationship between ⁵⁹Co and ⁵⁷Fe Chemical Shifts," Hagen, K. I.; Schwab, C. M.; Edwards, J. O.; Sweigart, D. A. Inorg. Chem. 1986, 23 , 978. "⁵⁹Co NMR of Six-Coordinate Cobalt(III) Tetraphenylporphyrin Complexes. 4. The Effect of Phenyl Ortho Substituents on Chemical Shifts, Line Widths, and Structure," Bang, H.; Edwards, J. O.; Kim, J.; Lawler, R. G.; Reynolds, R.; Ryan, W. J.; Sweigart, D. A. J. Am. Chem. Soc. 1992, 114, 2843.

14. "⁵⁹Co NMR Studies of Diamagnetic Porphyrin Complexes in the Solid Phase," Medek, A.; Frydman, V.; Frydman, L. J. Phys. Chem. B 1997, 101, 8959.