Jun 30, 1984 - (1) Sherman, S. E.; Lippard, S. J. Chem. Rev. 1987, 87, 1153-1181. (2) Loehrer, P. J.; Einhorn, L. H. Annal. Int. Med. 1984, 100, 704-713.
Ligand Loss Photochemistry of Ruthenium Complexes A Thesis Presented in Partial Fulfillment of the Requirements for the Degree Master of Science in the Graduate School of the Ohio State University
By Mark Sgambellone, B.S. Graduate Program in Chemistry
The Ohio State University 2009
Thesis Committee: Claudia Turró, advisor Yiying Wu
ABSTRACT The photo-induced ligand loss of the complexes [Ru(tpy)(AN)3]2+ and cis[Ru(tpy)(AN)2Cl]+ (tpy = 2,2’:6’2’’-terpyridine) was studied in water and in CH2Cl2 in the presence of chloride from tetrabutylammonium chloride (TBACl). Both complexes photolyze to the same photoproduct. Photolysis in CH2Cl2 in the presence of chloride ions led to the photoproduct trans-[Ru(tpy)(AN)Cl2], and photolysis in water led to the photoproduct trans-[Ru(tpy)(AN)(H2O)2]2+. The two axial acetonitrile ligands were replaced, while the equatorial acetonitrile remained coordinated to the metal. For cis-[Ru(tpy)(AN)3]2+ the axial acetonitrile ligands were replaced in a step-wise fashion, forming an intermediate with one axial acetonitrile. Polypyridyl ruthenium complexes such as cis-[Ru(bpy)2(AN)2]2+ are known to form the diaqua species cis-[Ru(bpy)2(H2O)2]2+ and bind to DNA upon photolysis, but not in the absence of light. Since [Ru(tpy)(AN)3]2+ and cis[Ru(tpy)(AN)2Cl]+ exhibit analogous photoreactivity in water, they are potential anti-tumor agents for use in photodynamic therapy (PDT). cis-[Ru(tpy)(AN)2Cl]+ has a lower energy metal to ligand charge transfer (MLCT) tranisiton and a higher quantum yield of ligand substitution, making it the better candidate for PDT. The series of complexes cis-[Ru(bpy)2L2]2+ (bpy = 2,2’-dipyridly) (L = NCPh, 4-F-NCPh, 4-Me-NCPh, and 4-OMe-NCPh) were synthesized and their ii
photolysis rate in water was studied. The complexes with more electron withdrawing substituents exhibited faster photolysis rates. Electronic absorption and electrochemical data showed π-back bonding in the complexes, with more πback bonding correlating with a faster rate of photolysis. The electrochemical data showed similar degrees of π-back bonding for cis-[Ru(bpy)2(NCPh)2]2+ and cis[Ru(bpy)2(4-Me-NCPh)2]2+, indicating that both amounts of σ bonding and π-back bonding play a role in the rate of photolysis. However, Cis-[Ru(bpy)2(4-OMeNCPh)2]2+ did not fit the trend.
iii
DEDICATION
Dedicated to Dr. Ignacio “Doc Oc” Ocasio.
iv
ACKNOWLEDGEMENTS
I would like to thank my advisor, Dr. Claudia Turro, for all of her help and guidance in my research and in writing this thesis. I would like to thank the members of the Turro research group (2007-2009) for their assistance as well.
v
VITA June 30th 1984……………………Born: Gainesville, Florida May 2003…….……………..........Highland High School May 2007………...........................BS Chemistry, Case Western Reserve University 2007 to present…………………..Graduate Teaching and Research Associate, Department of Chemistry, The Ohio State University
FIELDS OF STUDY Major Field: Chemistry
vi
TABLE OF CONTENTS
Page
Abstract………………………………………………………………………….ii Dedication……………………………………………………………………….iv Acknowledgements……………………………………………………………...v Vita………………………………………………………………………………vi List of Figures……………………………………………………………………ix List of Tables…………………………………………………………………….xii Chapters 1. Introduction………………………………………………………………….1 2. Background…………………………………………………………………..5 3. Expermimental…………………………………………………………….....13 Materials…………………………………………………………………13 Synthesis and Characterization…………………………………………..13 Instrumentation…………………………………………………………..18 Methods………………………………………………………………….19
vii
4. Ruthenium Terpyridine Compounds Absorption of Ru(tpy)XYZ compounds…………………………..……..28 Photoanation of Complexes CH2Cl2……………………………………..29 [Ru(tpy)(AN)3]2+…………………………………………………29 Cis-[Ru(tpy)(AN)2Cl]+…………………………………………...32 Photoaquation………………………………………………………........34 [Ru(tpy)(AN)3]2+……………………………………..…………..35 Cis-[Ru(tpy)(AN)2Cl]+……………………………………...........38 Quantum Yields………………………………………………..………...40 Dark Reactions……………………………………………………...........42 5. Ruthenium Bis-dipyridyl Benzonitrile Compounds Electronic Absorption Spectra…………………………………………...45 Photolysis in Water………………………………………………………47 Cis-[Ru(bpy)2(NCPh)2]2+ and cis-[Ru(bpy)2(4-Me-NCPh)2]2+…48 Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ and cis-[Ru(bpy)2(4-F-NCPh)2]2+…………………………………...51 Rates of Photolysis………………………………………………54 Electrochemistry………………………………………………………....57 Dark Reactions…………………………………………………………..59 6. Conclusion…………………………………………………………………...62 List of References………………………………………………………………..65
viii
LIST OF FIGURES
Figure
Page
1.1
Formation of diaqua cisplatin species………………………………..….1
1.2
Jablonski diagram for production of singlet oxygen………….………....3
1.3
Schematic representations of the structures of porphyrin, phthalocyanine and naphthalocyanine………..………………………….4
2.1 Schematic representation of cis-[Rh2(µ-O2CCH3)2(CH3CN)4(H2O)2]2+…….6 2.2 Structure of 2-(phenylazo)pyridine and 2,2’-azobis(pyridine)………………8 2.3 Jablonski diagram showing the general mechanism of photosubstitution of metal complexes……………………………………….9 2.4 UV-VIS spectra of Ru(bpy)32+ and Ru(tpy)22+ in acetonitrile……………....11 2.5 Structure of the series of Ru(bpy)L2 compounds synthesize…………….….12 3.1 General diagram for synthesis of Ru(tpy)Cl3………………...……………...14 3.2 General diagram for synthesis of Ru(tpy)(AN)3-nCln n=0-2……………...…14 3.3 Molecular structure with proton numbering scheme and 1
H NMR spectrum of trans-Ru(tpy)(AN)Cl2 in CDCl3...............………………21
3.4 Molecular structure with proton numbering scheme and 1
H NMR spectrum of Ru(tpy)(AN)2Cl in CD2Cl2…………………………..22
3.5 Molecular structure with proton numbering scheme and ix
1
H NMR spectrum of Ru(tpy)(AN)3 in d6-acetone…………………………..23
3.6 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(NCPh)2]2+ in d6-acetone………………………………………24 3.7 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-Me-NCPh)2]2+ in d6-acetone……..………………………....25 3.8 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ in d6-acetone…...…………………………26 3.9 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-F-NCPh)2]2+ in d6-acetone………………….………………27 4.1 Absorption of [Ru(tpy)(AN)3]2+, cis-[Ru(tpy)(AN)2Cl]+ and trans-Ru(tpy)(AN)Cl2 in CH2Cl2………………….………………….…29 4.2 Photolysis of Ru(tpy)(AN)3 in CH2Cl2 in the presence of chloride ions…….32 4.3 Photolysis of Ru(tpy)(AN)2Cl in CH2Cl2 in the presence of chloride ions….33 4.4 Electronic absorption spectrum of the photolysis products of [Ru(tpy)(AN)3]2+ and cis-[Ru(tpy)(AN)2Cl]+ photolyzed with Cl- compared with the spectrum of trans-Ru(tpy)(AN)Cl2 synthesized from a literature procedure……………………………………...34 4.5 Photolysis of [Ru(tpy)(AN)3]2+ in water……………………………………..35 4.6 Photolysis of Ru(tpy)(AN)32+ in D2O followed by NMR……………………37 4.7 NMR spectra of Ru(tpy)(AN)32+ in D2O and photoproduct after photolysis for 24 hours in D2O………………………………………………………….38 4.8 Photolysis of [Ru(tpy)(AN)2Cl]+ in water……………………….………...39 photolysis products of Ru(tpy)(AN)32+
4.9 Comparison of NMR spectra of x
in D2O and Ru(tpy)(AN)2Cl+ in D2O……………………….……………….40 4.10 Concentration of [Ru(tpy)(AN)3]2+ vs. irradiation time. Photolysis was done in CH2Cl2 in the presence of Cl-...........................................................41 4.11 Dark reaction of [Ru(tpy)(AN)3]2+ and cis-[Ru(tpy)(AN)2Cl]+ in water…...43 4.12 Dark reaction of [Ru(tpy)(AN)3]2+ and cis-[Ru(tpy)(AN)2Cl in CH2Cl2 with TBACl…………………………………………………………………44 5.1 Absorption spectra of cis-[Ru(bpy)2(NCPh)2]2+, cis-[Ru(bpy)2(4-Me- NCPh)2]2+, cis-[4-OMe-Ru(bpy)2(NCPh)2]2+ and cis-[Ru(bpy)2(4-F-NCPh)2]2+ in water…………………………………..46 5.2 Photolysis of cis-[Ru(bpy)2(NCPh)2]2+ in water……………………………..49 5.3 Photolysis of cis-[Ru(bpy)2(4-Me-NCPh)2]2+ in water………………………50 5.4 Photolysis of cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ in water………………….....52 5.5 Photolysis of cis-[Ru(bpy)2(4-F-NCPh)2]2+ in water………………………...53 5.6 Absorption maxima in water vs. rate of photolysis in water of the series of compounds cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, 4-F-NCPh)…….……………………………...57 5.7 Dark reaction of cis-[Ru(bpy)2(NCPh)2]2+ and cis-[Ru(bpy)2(4-Me-NCPh)2]2+ in water……..……………………………...60 5.8 Dark reaction of cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ and cis-[Ru(bpy)2(4-F-NCPh)2]2+ in water……………………….………………61
xi
LIST OF TABLES
Page
Table 2.1 Comparison of inhibitory dose of 50% (ID50, µM/L) of
Ru(azp)2Cl2 compared with cisplatin………………………..………………7 2.2 Quantum yields of photoanation with tetrabutylammoniumchloride (TBACl) in CH2Cl2……………………………………………..…………..12 4.1 Absorption maximum and extinction coefficients of Ru(tpy)L3 compound…………………………………………………………….……..29 4.2 Quantum yields of photoaquation and photoanation. λirr = 400 nm………...41 5.1 Absorption maximum and extinction coefficients of cis-[Ru(bpy)2L2]2+ compounds in water…………………………………….46 5.2 Rates of photolysis in water of the cis-[Ru(bpy)2L2]2+ series (λirr = 400 nm)……………………………………………………………….55 5.3 Half-wave potentials E1/2 (V) vs NHE of complexes [Ru(bpy)2L]2+ (L = AN, 4-F-NCPh, NCPh, 4-Me-NCPh, 4-OMe-NCPh, and bpy)………..58
xii
CHAPTER 1
INTRODUCTION
Since the discovery of the antitumor drug cis-diamminedichloroplatinum(II) (cisplatin), it and its analogs have been used to treat various types of cancer.1,2 The major mechanism by which cisplatin kills cancer cells is by binding to DNA and inhibiting cell replication.1,3,4 Upon entering the cell, water displaces the chloride ligands of cisplatin forming the cis-diamminediaquaplatinum(II) ion (Figure 1.1). The diaqua species binds to DNA primarily by forming 1,2intrastrand guanine-guanine bridges by coordinating at the N7 position of adjacent bases3-5 This cross link kinks the DNA, causing a bend of ~35-40 degrees towards the major groove.6 Proteins with high mobility group (HMG) domains bind to the 1,2-intrastrand d(GpG) platinum adduct preventing its excision by the enzyme excinuclease.7-9 This prevents cells from repairing themselves and increases the toxicity of the drug.
+ NH3
Cl
H2O
NH3
NH3
OH2
Cl
NH3
NH3
H2O
Pt
Pt
+2
Cl
Figure 1.1 Formation of diaqua cisplatin species. 1
OH2
Pt NH3
OH2
Two major drawbacks of cisplatin and its analogs are its toxicity towards healthy cells and acquired resistance.10,11 Cisplatin kills healthy cells along with cancer cells, and this limits the possible dosage of the drug, decreasing its effectiveness. Second and third generation cisplatin complexes have been designed with lower toxicity, but all of these drugs are still toxic to healthy cells as well as cancerous cells. This is because cisplatin and its analogs are not selective for cancer cells. They are thermally activated exchanging to the diaqua species upon entrance into the cells, and this will necessarily happen in healthy cells as well as cancerous cells. Another problem is intrinsic and acquired resistance to cisplatin and its analogs.10-12 Cells with a high degree of excision repair tend to be resistant to cisplatin. A drug that is photo-activated, and not thermally activated, would have a significant advantage. Side effects can be reduced by only irradiating tumors, and this may allow for higher dosages making resistance less of a problem. Treatment of cancer by photo-activated drugs is generally known as photodynamic therapy (PDT). A good PDT drug is one that has a high quantum yield of photoaquation upon irradiation of light, absorbs strongly in the low energy visible (red) region of the electromagnetic spectrum (650-850nm), and possesses minimal dark toxicity. Human tissue is relatively transparent in this wavelength range, and a compound that absorbs light in this range would have maximum penetration inside the body.13 One class of PDT drugs that has been used successfully is porphyrins. Upon irradiation these compounds are excited to a singlet state, which can convert to a long lived triplet excited state through 2
inter-system crossing (Figure 1.2). As shown in Figure 1.2, the triplet can undergo energy transfer producing singlet oxygen, 1O2, which is highly reactive towards biomolecules and damages cellular components. Photofrin is a PDT drug approved by the FDA that has been used for the treatment of lung cancer, oesophagal cancer, and bladder cancer among others.13 Photophrin is composed of porphyrin derivatives (Figure 1.3). Although porphyrins have been effective in treating certain types of cancer, its lowest energy absorption band is at 630
Figure 1.2 Jablonski diagram for production of singlet oxygen, showing (a) absorption of light (b) Fluorescence (c). Internal conversion (d) Intersystem Crossing (e) Phosphorescence and (f) Singlet Oxygen Production.
nm, and it is not very strong. In order to increase the wavelength and extinction coefficient of absorption, the π-systems of the porphyrins were extended in
3
phthalocyanine and naphthalocyanine rings (Figure 1.3). As shown in Figure 1.3, metal atoms can be inserted inside of the rings, and axial ligands can be used to tune the solubility of the compounds. Naphthalocyanines absorb wavelengths ~770 nm.13 The major disadvantage of this type of PDT agent is the requirement of oxygen for the action of the drug and result of cell death. Cancer cells are usually hypoxic, which makes a mechanism requiring oxygen less than ideal. A possible means to address these drawbacks is a cisplatin analogue in which the species does not undergo conversion to the reactive diaqua species until it is irradiated with light. This type of PDT agent would not depend on oxygen and could have improved toxicity in cancer cells which are often hypoxic.
a)
(b)
(c) R R
N N
N
N
N
NH
N
M
N
N
M N
N
NH
N
N
N
N N
N
N
N
R
R
Figure 1.3 Schematic representations of the structures of (a) porphyrin, (b) phthalocyanine and (c) naphthalocyanine.
4
CHAPTER 2
BACKGROUND
Two metals that have been studied extensively for use as PDT agents are rhodium14-25 and ruthenium.26-34 Complexes of these metals have been shown to cause cell death through either singlet oxygen production or oxygen independent damage and DNA binding upon irradiation with light. Some of these complexes show improvements over cisplatin and Photofrin in cell toxicity and in the quantum yield of production of singlet oxygen. One of the characteristics of a good PDT drug is absorbance in the near infrared spectral region. Dirhodium species of the type Rh2(O2CCH3)4 (Figure 2.1) have low energy metal centered excitation of the rhodium-rhodium bond, making them a good candidate for PDT.19 A dirhodium cisplatin analogue, which loses ligands upon irradiation with light and binds to DNA, is cis-[Rh2(µO2CCH3)2(CH3CN)4(H2O)2]2+ whose structure is shown in Figure 2.1. Upon irradiation with light, two of the CH3CN ligands are replaced by water, forming a complex
that
binds
to
DNA.
The
cytotoxicity
of
cis-[Rh2(µ-
O2CCH3)2(CH3CN)4(H2O)2]2+ increases by a factor of 34 upon irradiation,14 which is significantly greater than the 5 fold increase measured for the key 5
component of the drug Photofrin under the same experimental conditions.
2+
CH3 O
O
CH3
Rh
O OH2
O H2O
AN
Rh AN AN
AN
Figure 2.1 Schematic representation of cis-[Rh2(µ-O2CCH3)2(CH3CN)4(H2O)2]2+.
Ruthenium complexes with labile ligands have been studied extensively as potential pharmaceuticals.37 There are several mechanisms by which ruthenium can undergo ligand loss and subsequent binding to DNA including thermal ligand loss, activation by reduction, and photoinduced ligand loss. One compound known to react thermally with DNA is Ru(azp)2Cl2 (azp = 2-(phenylazo)pyridine) shown in Figure 2.2. Once inside the cell, Ru(L)XCly complexes can hydrolyze to Ru(L)x(H2O)y complexes in a similar way as cisplatin.38,39 Ruthenium compounds with bidentate aromatic imine ligands such as azp and bpy have better antitumor activity than other ruthenium chlorides such as cis-Ru(DMSO)4Cl2.38,39 A possible reason is that the π-acceptor effect of the imines decreases the rate of aquation to the aqua species to a level that is similar to cisplatin.37 Increased hydrophobic interactions between the aromatic ligands and DNA may also be responsible, as well as geometric effects of the ligands effecting protein binding to the DNA.37 Ruthenium diaqua species such as Ru(azp)2(H2O)22+ and Ru(bpy)2(H2O)22+ (bpy = 2,2’-dipyridly) are known to bind to DNA base adducts and show cytotoxicity, 6
presumably by binding to DNA.31,37-40
Table 2.1 shows the ID50 values for
Ru(azp)2Cl2, which undergoes thermal aquation to Ru(azp)2Cl2, compared with other metal complexes. Ru(azp)2Cl2 shows cell toxicity better than that of cisplatin in several cases.
Cell line
MCF-7
EVSA-T WIDR
IGROV
M19
A498
H266
ID50 Ru(azp)2Cl2
0.6
0.1
1.9
0.8
0.2
1.2
1.5
ID50 Cisplatin
2.3
1.4
3.2
0.6
1.9
7.5
10.9
Table 2.1 Comparison of inhibitory dose of 50% (ID50, µM/L) of Ru(azp)2Cl2 compared with cisplatin.38
Binding to a model DNA bases such as 9-ethlyguanine has been observed in several ruthenium compounds with aromatic imine ligands including Ru(bpy)2(H2O)22+,
Ru(phen)2(H2O)22+
(phen
=
2-phenylpyridine)
and
Ru(tpy)(bpy)(H2O)2+ (tpy = 2,2’:6’,2”-terpyridine).40 The cytotoxicity in cells of the series Ru(bpy)2Cl2, Ru(bpy)(azp)Cl2 and Ru(azp)2Cl, which produce the above diaqua species upon irradiation, was studied and found to vary from the inactive Ru(bpy)2Cl2 to the highly cytotoxic Ru(azp)2Cl2 with Ru(azp)(bpy)Cl2 showing intermediate cytotoxicity.32 The mechanism of cell death is thought to proceed through the diaqua species, and the difference in cytotoxicity may be due to the different electronic properties of azp compared to the bpy ligand. Due to the high antitumor activity of Ru(tpy)Cl341, the binding of Ru(tpy)(apy)(AN) and Ru(tpy)(apy)Cl (apy = 2,2’-azobis(pyridine) Figure 2.2) to the DNA model base 7
9-ethylguanine has also been studied.30 Both compounds hydrolyzed to the aqua species and bound to 9-ethylguanine, with Ru(tpy)(apy)(AN) showing much faster binding.
(a)
(b)
N
N
N
N
N
N
N
Figure 2.2 Structure of (a) 2-(phenylazo)pyridine and (b) 2,2’-azobis(pyridine)
Reduction of inert Ru(III) compounds activates them by their conversion to Ru(II) in vivo, which can then thermally bind to DNA. In general, upon reduction to Ru(II), π- donating ligands become destabilized and hydrolyze. Two compounds that have been shown to operate by this mechanism are cis[RuCl2(NH3)4]Cl and [ImH]trans-[Ru(Im)2Cl4] (Im = imidazole). Activation by reduction has the advantage that tumor cells are generally more acidic than healthy cells and as a result have a lower electrochemical potential relative to healthy cells, making their interior more reducing.37 This difference between healthy and cancerous cells lends partial selectivity to these compounds. Phototinduced ligand loss to the diaqua species represents another 8
manner to achieve selective toxicity. Ruthenium complexes have high intensity metal to ligand charge transfer (MLCT) absorbances and long lived excited states making them good candidates for PDT. Photosubstitution often occurs from a ligand field state (LF) that is thermally accessible from the 3MLCT.47 Excitation with visible light can excite an electron to the 1MLCT which can access the 3
MLCT through internal conversion. Second and third row transition metals
undergo internal conversion fairly well due to spin-orbit coupling. The processes are shown in Figure 2.3. Ru(tpy)(4-CO2H-4’-Mebpy)(NO2)+, Ru(bpy)2(NH3)22+, and Ru(bpy)2(AN)22+ are all known to undergo substitution to form the aqua species upon irradiation with light.26,29,35 Ru(tpy)(4-CO2H-4’-Mebpy)(NO2)+ has an MLCT transition at 475 nm, and has been observed upon irradiated to produce Ru(tpy)(4-CO2H-4’-Mebpy)(MeCN)+ in acetonitrile, and was shown to react with calf thymus DNA in aqueous solution upon irradiation for 30 min.29
Figure 2.347 Jablonski diagram showing photosubstitution of metal complexes. 9
the
general
mechanism
of
Ru(bpy)2(NH3)22+ has a low energy MLCT at 490 nm (ε = 8,210 M-1cm-1) and is known to lose the NH3 ligands upon irradiation.26 When irradiated at 400 nm in water, the NH3 ligands are substituted with water molecules to form cisRu(bpy)2(H2O)22+ with a quantum yield of Φ = 0.018.35 The compound has been shown to bind to the 15-mer single stranded oligonucleotide sequences 5’TGCAAGCTTGGCACT-3’ and 5’- AGTGCCAAGCTTGCA-3’, and to double stranded DNA when irradiated at > 345 nm.26 Cis-Ru(bpy)2(CH3CN)22+ has a higher energy MLCT transition with a maximum at 425 nm (ε = 8,900 M-1cm-1). When Irradiated at 400 nm in water, it has been shown to undergo photoaquation in water with a quantum yield of Φ = 0.21,35,49 and to bind to double stranded DNA when irradiated at >345 nm.35 The two complexes acted very similar towards DNA, the only major differences between them
is
that
[Ru(bpy)2(NH3)2]2+
absorbs
at
a
lower
wavelength
and
[Ru(bpy)2(CH3CN)2]2+ has a higher quantum yield of photoaquation.
Two important features of a photodynamic therapy agent are a low energy MLCT transition and a high quantum yield of photoaquation. Low energy MLCT transitions are desirable because organic tissue becomes more transparent in the near infrared wavelengths. The increased π-conjugation of the 2,2’:6’,2’’terpyridine (tpy) results in a lower MLCT transition as compared to 2,2’-dipyridyl (bpy), as shown in Figure 2.4. For this reason, Ru(tpy)L3 complexes were chosen for study as possible PDT agents. The ligand, L, was chosen for maximum quantum yield of photo substitution in order to get maximum conversion to the activated aqua species. Comparing known Ru(bpy)2L2 complexes, the highest 10
quantum yield of photoanation is when the ligand, L, is acetonitrile.35,36 Table 2.2 compares the quantum yield of photoanation with chloride of a series of Ru(bpy)2L22+ compounds. Acetonitrile is thought to have a large quantum yield of ligand loss because it binds to the metal center through both weak σ and π-back bonding. Excitation of the MLCT transition involves transfer of an electron from the metal to one of the bpy ligands, reducing the electron density on the metal and destabilizing the π-back bonding to acetonitrile. For these reasons, Ru(tpy)(AN)3, Ru(tpy)(AN)2Cl, and Ru(tpy)(AN)Cl2 were studied as possible agents for photodynamic therapy. Photolysis of these compounds causes photosubstitution of the acetonitrile ligands forming an aqua species in water that should bind to DNA bases.
Figure 2.4 UV-VIS spectra of Ru(bpy)32+ (solid line) and Ru(tpy)22+ (dashed line) in acetonitrile.
11
Complex Quantum yielda 2+ Ru(bpy)2(AN)2 0.31 mol/Einstein Ru(bpy)2(CO)22+ 0.05 mol/Einstein 2+ Ru(bpy)2(py)2 0.20 mol/Einstein Table 2.2 Quantum yields of photoanation with tetrabutylammoniumchloride (TBACl) in CH2Cl2. (A) taken from reference 36. An investigation into the large quantum yield of photosubstitution of acetonitrile was also carried out. The a series of Ru(bpy)L2 compounds, shown in Figure 2.5, were synthesized and photolyzed in water. Since the large quantum yield is thought to be due to disruption of π-back bonding to the metal upon excitation of the MLCT, varying the functional group at the para position of the coordinated benzonitrile, and therefore the electron density on the ligand, should affect the rate of photosubstitution. Adding electron withdrawing groups should increase the back bonding character of the metal ligand bond, causing an MLCT transition to have a larger effect on the strength of the ruthenium benzonitrile bond. This would correlate as an increase in photoinduced ligand loss. An electron donating group should reduce the back bonding character of the metal ligand bond, causing the photoinduced ligand loss to decrease.
N
N
N
N
N N
N
L=
Ru L
N L
O
Figure 2.5 Structure of the series of Ru(bpy)L2 compounds synthesized 12
F
CHAPTER 3
EXPERIMENTAL
3.1 Materials 3.1.2 Commercial Materials Water used in all reactions and measurements was deionized using Barnstead Fi-stream filter system to 18 MΩ. H2SO4, Et2O, and N(Et)3 purchased from Fisher, CHCl3 and MeOH purchased from Mallinckrodt, and CH3CN and EtOH absolute purchased from Acros, and Decon labs, respectively were used without further purification. RuCl3 trihydrate, Bu4NPF6, NH4PF6, 2,2’-dipyridyl, N,N-dimethylformamide, and 2,2’:6’,2”-terpyridine were purchased from Aldrich, Bu4NCl and 1,10-phenanthroline were purchased from Fluka, LiCl was purchased from J. T. Chemical Co. and used as received. Sodium acetate trihydrate was purchased from Sigma, and (NH4)3[Fe(C2O4)3] trihydrate was purchased from Riedel-de Haën and used as received. 3.1.2 Synthesis and Characterization All compounds were prepared from Ru(tpy)Cl3 which was obtained from RuCl3*3H2O as shown in Figure 3.1.
13
Cl N N
RuCl3
EtOH absolute, reflux 3 h
N
N
Ru N
N
Cl Cl
Figure 3.1 General diagram for synthesis of Ru(tpy)Cl3
Ru(tpy)Cl3 was used as a starting reagent for a series of Ru(tpy)(AN)3-nCln compounds as shown in Figure 3.2. Cl
N
CH3CN, Et3N,
N
Ru
CHCl3 reflux 1.5 h
N NCCH3 Cl
Cl
Cl
N
N
CH3CN
N
N
EtOH absolute/water reflux 9 h
N
Ru
Ru N
NCCH3
Cl NCCH3
Cl
NCCH3
CH3CN N
EtOH absolute/water reflux 9 h
N
Ru N NCCH3 NCCH3
Figure 3.2 General diagram for synthesis of Ru(tpy)(AN)3-nCln n=0-2
14
The details of the synthesis of each complex are described below. Ru(tpy)Cl3. Ru(tpy)Cl3 was synthesized by a method previously reported.43 A mixture of 476.2 mg (1.82 mmol) RuCl3·3H2O, 399.9 mg (1.71 mmol) 2,2’:6’2’’-terpyridne, and 130 mL of absolute ethanol was stirred under reflux for 3 hours. The solution was filtered and washed with three 30 mL portions of absolute ethanol to remove Ru(tpy)2+, three 30 mL portions of diethyl ether to remove excess tpy, and air dried. Yield: 606 mg (80%). 1H NMR (400 MHz, CD3CN) δ(ppm): 8.3 (d, 2H), 8.2 (m, 2H), 7.8 (m, 2H), 7.7 (t, 2H), 7.6 (t, 3H). Ru(tpy)(AN)Cl2. Ru(tpy)(AN)Cl2 was prepared according to a literature procedure.44 A mixture of 122.7 mg (0.278 mmol) Ru(tpy)Cl3, 1 mL acetonitrile, and 1 mL of triethyl amine was refluxed for 1.5 hours in 30 mL of chloroform. The solution was cooled to room temperature, filtered, and 30 mL of ethanol was added. The solution was condensed to ~20 mL, filtered, and air dried. Yield: 96.7 mg (78%). 1H NMR (250 MHz, CDCl3) δ(ppm): 9.17 (d, J= 5.82 Hz, 2H), 8.05 (dd, J = 13.30, 7.75 Hz, 4H), 7.78 (t, J=7.70 Hz, 2H), 7.63 (t, J=7.90 Hz, 1H), 7.46 (t, J=7.05 Hz, 2H), 2.85 (s, 3H) shown in Figure 3.3. [Ru(tpy)(AN)2Cl](PF6). Ru(tpy)(AN)2Cl was prepared following a literature procedure.44 A mixture of 138.9 mg (0.315 mmol) Ru(tpy)Cl3 and 3 mL acetonitrile was refluxed for 9 hours in 1:1 (v/v) ethanol : water. The solution was condensed to ~10 mL and NH4PF6(aq) was added. The solution was filtered and chromatographed on an alumina I (basic) column using 1:1 (v/v) dichloromethane/acetone as the
eluent. Yield: 21.6 mg (11%). 1H NMR 15
(400 MHz, acetone) δ(ppm): 9.12 (ddd, J = 5.47, 1.51, 0.76 Hz, 2H), 8.57 ( m, 4H), 8.13 ( m, 3H), 7.74 (ddd, J = 7.60, 5.46, 1.32 Hz, 2H), 2.94 (s, 3H), 2.13 (s, 3H) shown in Figure 3.4. [Ru(tpy)(AN)3](PF6)2. Ru(tpy)(AN)3 was prepared following a literature procedure.44 A mixture of 138.9 mg (0.315 mmol) Ru(tpy)Cl3 and 3 mL acetonitrile was refluxed for 9 hours in 1:1 (v/v) ethanol : water. The solution was condensed to ~10 mL and NH4PF6(aq) was added. The solution was filtered and chromatographed on an alumina I (basic) column using 1:1 (v/v) dichloromethane/acetone as the eluent. Yield: 36.2 mg (15%). 1H NMR (400 MHz, acetone) ppm 9.15 (dd, J = 5.42, 0.72 Hz, 2H), 8.69 (dd, J = 8.07, 1.74 Hz, 4H), 8.33 ( m, 3H), 7.87 (ddd, J = 7.62, 5.46, 1.26 Hz, 2H), 2.92 (s, 3H), 2.16 (s, 3H) shown in Figure 3.5. The chloride salts [Ru(tpy)(AN)3]Cl2 and [Ru(tpy)(AN)2Cl]Cl were prepared by adding an acetone solution of N(Bu)4Cl to the PF6 salt of each compound dissolved in acetone. The solutions were filtered and the solid product was air dried. Cis-Ru(bpy)2Cl2. Cis-Ru(bpy)2Cl2 was synthesized by a method previously reported.50 A mixture of 891.2 mg (1.84 mmol) RuCl3·3H2O, 1.0266 g (6.57 mmol) 2,2’-dipyridyl and 1.0799 g (25.5 mmol) LiCl was refluxed for 8 hours under nitrogen in 2:5 (v/v) N,N-dimethylformamide/methonal. 35 mL of acetone was added and the solution was cooled in a freezer overnight. The solid was filtered and dissolved in CH2Cl2 and [Ru(bpy)3]2+ was extracted with water until no essentially no [Ru(bpy)3]2+ could
be seen in the water layer. The CH2Cl2 16
was evaporated and the solid collected. Yield: 630.3 mg (40%). Shown in Figure 3.6 Cis-[Ru(bpy)2(NCPh)2]2+. Cis-[Ru(bpy)2(NCPh)2]2+ was synthesized by a method previously described.51 A mixture of 18.5 mg (0.0382 mmol) cisRu(bpy)Cl2 and 4 mL (4.38 mmol) benzonitrile was refluxed under nitrogen in 1:1 (v/v) methonal/water for 2 hours. The methonal was evaporated and excess benzonitrile ligand was extracted with 3X30 mL ether. The product was precipitated by addition of NH4(PF6) and collected by filtration. The compound was chromatographed on an alumina I (basic) column using 1:1 (v/v) dichloromethane/acetone as the eluent. The second fraction was collected and dissolved in minimal acetone and precipitated by addition of ether. Yield: 3.6 mg (10%). 1H NMR (400 MHz, d6-acetone) ppm 9.80 (d, J = 5.05 Hz, 2H), 8.89 (d, J = 8.10 Hz, 2H), 8.75 (d, J = 8.06 Hz, 2H), 8.48 (dt, J = 8.01, 7.98, 1.24 Hz, 2H), 8.18 (dt, J = 8.10, 8.01, 1.21 Hz, 2H), 8.06 (dd, J = 12.62, 6.23 Hz, 4H) 7.84-7.72 (m, 6H), 7.58 (t, J = 7.87 7.87 Hz, 4H) 7.55-7.46 (m, 2H), shown in Figure 3.7 Cis-[Ru(bpy)2(4-Me-NCPh)2]2+. Cis-[Ru(bpy)2(4-Me-NCPh)2]2+ was synthesized by a procedure analogous to that of Cis-[Ru(bpy)2(NCPh)2]2+ except 4-methylbenzonitrile was used in place of benzonitrile. Yield: 17.1 mg (17%). 1H NMR (400 MHz, d6-acetone) ppm 9.77 (dd, J = 4.92, 0.70 Hz, 2H), 8.88 (d, J = 8.04 Hz, 2H), 8.74 (d, J = 7.95 Hz, 2H), 8.47 (t, J = 7.92, 7.92 Hz, 2H), 8.16 (t, J = 7.87, 7.87 Hz, 2H), 8.04 (dd, J = 12.93, 5.99 Hz, 4H), 7.64 (d, J = 8.10 Hz, 4H), 7.50 (t, J = 7.21, 7.21 Hz, 2H), 7.39 (d, J = 8.46 Hz, 4H), 2.40 (s, 6H) shown in Figure 3.8 17
Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+. Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ was synthesized by a procedure analogous to that of Cis-[Ru(bpy)2(NCPh)2]2+ except 4-methoxybenzonitrile was used in place of benzonitrile. Yield: 37.3 mg (18%). 1
H NMR (500 MHz, d6-acetone) ppm 9.73 (d, J = 5.14 Hz, 2H), 8.85 (d, J = 7.85
Hz, 2H), 8.71 (d, J = 8.01 Hz, 2H), 8.44 (t, J = 7.50, 7.50 Hz, 2H), 8.13 (t, J = 7.54, 7.54 Hz, 2H), 8.08-7.97 (m, 4H), 7.70 (d, J = 8.80 Hz, 4H), 7.48 (t, J = 6.44, 6.44 Hz, 2H), 7.06 (t, J = 8.36, 8.36 Hz, 4H), 3.87 (s, 6H) shown in Figure 3.9 Cis-[Ru(bpy)2(4-F-NCPh)2]2+. Cis-[Ru(bpy)2(4-F-NCPh)2]2+ was synthesized by a procedure analogous to that of Cis-[Ru(bpy)2(NCPh)2]2+ except 4fluorobenzonitrile was used in place of benzonitrile. Yield: 11.1 mg (23%). 1H NMR (400 MHz, d6-acetone) ppm 9.77 (d, J = 5.40 Hz, 2H), 8.86 (d, J = 8.16 Hz, 2H), 8.73 (d, J = 8.13 Hz, 2H), 8.46 (dt, J = 7.92, 7.87, 1.34 Hz, 2H), 8.16 (dt, J = 7.92, 7.86, 1.30 Hz, 2H), 8.08-7.97 (m, 4H), 7.89 (ddd, J = 7.82, 4.97, 2.34 Hz, 4H), 7.53-7.46 (m, 2H), 7.40-7.29 (m, 4H) shown in Figure 3.10. 3.2 Instrumentation 1
H NMR spectra were obtained on a 250, 400, or 500 MHz Bruker system.
Electronic absorption spectra were performed on a Hewlett-Packard diode Array Spectrometer (HP 8453) equipped with HP 8453 Win System software. Sample solutions were contained in a 1 cm quartz cell in air unless otherwise noted. Quantum yields were recorded using a 150W Xe arc lamp from Photon Technology International with an LPS-220 lamp power supply.
18
3.3 Methods 3.3.1 Quantum Yields Photolysis reactions were carried out on a 150 W Xe arc lamp. The light beam was filtered through a Newport 400 nm band pass filter model 10BPF10400 with a FWHM = 10 nm. The absorption of the compounds at the irradiation wavelength was ~0.8 for Ru(tpy)(AN)32+ and ~0.5 for Ru(tpy)(AN)2Cl+. The intensity of the incident light was measured by ferrioxalate actinometry. Ferric oxalate has a well characterized photochemical reaction through which the iron(III) is reduced to iron(II). The ligand 1,10-phenanthroline coordinates almost exclusively with iron(II) over iron(III) to form a metal complex, Fe(phen)32+, which absorbs in the visible region. This allows the reaction to be monitored by UV-Vis spectroscopy. The quantum yield of the iron oxalate reaction is known for a wide range of irradiation wavelengths, which allows the calculation of the intensity of the lamp via equation 1.45 All solutions were degassed by bubbling nitrogen for 10 minutes prior to photolysis. The quantum yield was calculated from equation 1,
Φ=
MV It
(1)
where M is the change in molarity of the reactant, V is the volume of the irradiated sample, I is the intensity of the lamp, and t is the irradiation time.
45
Once the intensity of the lamp is known, equation 1 can be used to calculate the quantum yield of any species. A graph of the concentration of the reactant vs. irradiation time was constructed, and
the quantum yield is determined from the 19
initial slope of the graph. The reactant concentration during the photolysis is determined by equation 2, the simultaneous solution of Beer’s law for the reactant and product species, where
ε P , 400 A549 − ε P ,549 A400 MR = ε P , 400ε R ,549 − ε P ,549ε R , 400
(2)
Ax is the absorbance at wavelength x, εp,x is the molar extinction coefficient of the product at wavelength x, and εr,x is the molar extinction coefficient of the reactant at wavelength x.
20
C D N
F
N
B A Cl
Ru
N
Cl
H
N
A D
B C
D A
9.50
9.00
8.50
E
C
F
8.00
B
7.50
7.00
ppm (t1)
E D CFB
A
10.0 ppm (t1)
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
Figure 3.3. Molecular structure with proton numbering scheme and 1H NMR spectrum of Ru(tpy)(AN)Cl2 in CDCl3. Assignments from reference 44.
21
C
B
D N
F
A Cl
N
Ru
N
N
H
N
A
G D
B C
E D A
9.50
C
9.00
8.50
B
8.00
7.50
G
ppm (t1)
H D D E
A
C
9.0
8.0
B
7.0
6.0
5.0
4.0
3.0
2.0
1.0
ppm (t1)
Figure 3.4. Molecular structure with proton numbering scheme and 1H NMR spectrum of Ru(tpy)(AN)2Cl in CD2Cl2. Assignments from reference 44.
22
C D N
F
N
B A G
N
Ru
N
N
H
N
A
G D
B C
D E
F C
A
9.00
8.50
B
8.00
ppm (t1)
E D A
9.0
F C B
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
ppm (t1)
Figure 3.5. Molecular structure with proton numbering scheme and 1H NMR spectrum of Ru(tpy)(AN)3 in d6-acetone. Assignments from reference 44. 23
N N
N
Ru
N
N
N
9.50
9.00
8.50
8.00
7.50
ppm (t1)
9.0
8.0
7.0
6.0
5.0
4.0
3.0
ppm (t1)
Figure 3.6 Molecular structure and 1H NMR spectrum of cis[Ru(bpy)2(NCPh)2]2+ in d6-acetone
24
2.0
1.0
N N
N
Ru
N
N
N
9.50
9.00
8.50
8.00
7.50
ppm (t1)
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
ppm (t1)
Figure 3.7 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-MeNCPh)2]2+ in d6-acetone 25
O
N N
N
Ru
N
N
O
N
9.50
9.00
8.50
8.00
7.50
8.0
7.0
6.0
5.0
ppm (t1)
9.0
4.0
3.0
2.0
1.0
ppm (t1)
Figure 3.8 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-OMeNCPh)2]2+ in d6-acetone 26
F
N N
N
Ru
N
N
F
N
9.50
9.00
8.50
8.00
7.50
ppm (t1)
9.0
8.0
7.0
6.0
5.0
4.0
3.0
2.0
1.0
ppm (t1)
Figure 3.9 Molecular structure and 1H NMR spectrum of cis-[Ru(bpy)2(4-FNCPh)2]2+ in d6-acetone
27
CHAPTER 4
Ruthenium Terpyridine Compounds
4.1 Absorption of Ru(tpy)XYZ compounds The
electronic
absorption
spectra
of
[Ru(tpy)(AN)3]2+
,
cis-
[Ru(tpy)(AN)2Cl]+, and trans-Ru(tpy)(AN)Cl2 have been previously reported, and are shown in Figure 4.1.40 The absorbance maxima and molar extinction coefficients for each compound are listed in Table 4.1. The absorption spectra of all three compounds show ligand centered 1ππ* transitions on the tpy ligand with maxima at ~300 nm. [Ru(tpy)(AN)3]2+ and cis-[Ru(tpy)(AN)2Cl]+ exhibit singlet metal to ligand charge transfer (1MLCT) transitions from the ruthenium t2g state to the π* orbital on the tpy ligand, with maxima at 434 nm and 485nm respectively. Trans-Ru(tpy)(AN)Cl2 has two 1MLCT transitions at 400 and 549 nm. Upon excitation of the 1MLCT the complex can undergo either fluorescence to the ground state or intersystem crossing to the 3MLCT. It has been previously reported that neither [Ru(tpy)(AN)3]2+ nor cis-[Ru(tpy)(AN)2Cl]+ exhibit roomtemperature emission.44 This could be due to solvent quenching, since emission of ruthenium terpyridine complexes has been related to the ability of solvent molecules to access pockets between ligands.60 It is from the 3MLCT state, or a 28
state that is thermally accessible from the 3MLCT state, that photo substitution occurs.47,48
FiFigure 4.1 Absorption of [Ru(tpy)(AN)3]2+ (solid line) cis-[Ru(tpy)(AN)2Cl]+ (dashed line) and trans-Ru(tpy)(AN)Cl2 (dotted line) in CH2Cl2. λmax, nm (ε, M-1cm-1) in λmax, nm (ε, M-1cm-1) in CH2Cl2a water 2+ [Ru(tpy)(AN)3] 434 (4400) 434 (4000) [Ru(tpy)(AN)2Cl]+ 485 (4600) 460 (3950) Ru(tpy)(AN)Cl2 400 (5300) NA 549 (4800) Table 4.1 Absorption maximum and extinction coefficients of Ru(tpy)L3 compounds. a, from ref 44. Complex
4.2 Photoanation of Complexes in CH2Cl2. 4.2.1 [Ru(tpy)(AN)3]2+ Irradiation of the 1MLCT transition of [Ru(tpy)(AN)3]2+ or cis-[Ru(tpy)(AN)2Cl]+ 29
initiates ligand loss of acetonitrile as previously reported.40 The acetonitrile ligand is replaced by either a coordinating anion in solution or a solvent molecule. The photolysis of 130 µM [Ru(tpy)(AN)3]2+ in the presence of chloride ions was followed by UV-Vis spectroscopy and is shown in Figure 4.2. The compound was irradiated with 400 nm light in dichloromethane with ~10 mM chloride ions from added tetrabutlyammonium chloride (TBACl). As the acetonitrile ligands are replaced by chloride, the absorbance of the 1MLCT state becomes red shifted. This is due to the decreased ligand field splitting of the chloride ligand compared to the acetonitrile ligand. The product of the photolysis is trans-Ru(tpy)(AN)Cl2, as is apparent from comparing the UV-Vis spectrum of photolyzed [Ru(tpy)(AN)3]2+ to that of synthesized trans-Ru(tpy)(AN)Cl2 (Figure 4.4). The presence of an intermediate can be seen in the photoanation of [Ru(tpy)(AN)3]2+ in Figure 4.2 by the formation a peak at ~485nm at 2 min, and its subsequent disappearance around 8 minutes. The lack of constant isobestic points throughout the photolysis also indicates an intermediate. The intermediate is identified as cis-[Ru(tpy)(AN)2Cl]+ by the location of the peak at ~485nm and as previously reported.44 The photolysis reaction is shown in equation 1, and has been previously described.44 Since the intermediate is a stable compound that can be isolated, the overall reaction is most likely a two photon process, requiring one photon for the formation of cis-[Ru(tpy)(AN)2Cl]+, and a second photon for the formation of trans-Ru(tpy)(AN)Cl2.
30
axial N
N N H3CCN
NCCH3 Ru
NCCH3
hv, Cl-
N H3CCN
N
N Cl
Ru
NCCH3
N
hv, Cl-
N
equitorial
Cl Ru
NCCH3
Cl N
(1)
The axial acetonitrile ligands are replaced by chloride ions, while the equatorial actonitrile ligand trans to the tpy ligand remains coordinated to the metal. The difference in reactivity of axial and equatorial acetonitrile ligands can be explained by the trans effect, both acetonitrile and chloride have a stronger trans effect than pyridine, and therefore the ligand trans to the terpyridine ring is less likely to be substituted. Figure 4.4 compares the absorption spectra of the photoproducts of Ru(tpy)(AN)3 and Ru(tpy)(AN)2Cl with that of synthesized Ru(tpy)(AN)Cl2. It can be seen that the photo product of both species is in fact Ru(tpy)(AN)Cl2. It is necessary to de-gas the solutions with nitrogen prior to photolysis. In the presence of air, a different photolysis product is obtained. No 1H NMR signal could be detected for this photolysis product and is therefore most likely an oxidized paramagnetic ruthenium(III) compound.
31
Figure 4.2 Photolysis of 130 µM Ru(tpy)(AN)3 in CH2Cl2 in the presence of ~10 mM chloride ions from TBACl. 150W lamp 400 nm band pass filter. 4.2.2 Cis-[Ru(tpy)(AN)2Cl]2+ The phototlysis of cis-[Ru(tpy)(AN)2Cl]2+ results in the loss of the axial acetonitrile ligand. The acetonitrile is replaced by either a coordinating anion in solution or a solvent molecule, as shown in equation 2. Figure 4.3 shows the photolysis of 250 µM cis- [Ru(tpy)(AN)2Cl]2+ in dichloromethane with ~10 mM
N
N N H3CCN
Cl Ru
N
NCCH3
Cl
-
hv, Cl
N
Ru
NCCH3
Cl N
(2)
TBACl. The compound was irradiated at 400 nm, and the photolysis product was trans-Ru(tpy)(AN)Cl2 as previously reported44 and characterized by UV-Vis 32
spectroscopy. Figure 4.4 compares the photoproduct of cis-[Ru(tpy)(AN)2Cl]+, photolyzed in the presence of Cl-, to that of synthesized trans-Ru(tpy)(AN)Cl2. The axial ligand is replaced while the equatorial acetonitrile ligand remains bound to the metal. The difference in the reactivity of the two acetonitrile ligands can also be explained by the trans effect analogously to the case of [Ru(tpy)(AN)3]2+ described above.
Figure 4.3 Photolysis of 250 µM Ru(tpy)(AN)2Cl in CH2Cl2 in the presence of ~10 mM chloride ions from TBACl. 150W lamp 450 nm band pass filter.
33
Figure 4.4 Electronic absorption spectrum of the photolysis products of [Ru(tpy)(AN)3]2+ (dotted line) and cis-[Ru(tpy)(AN)2Cl]+ (dashed line) photolyzed in the presence of Cl- compared with the spectrum of transRu(tpy)(AN)Cl2 (solid line) synthesized from a literature procedure.44
4.3 Photoaquation Ru(tpy)(AN)32+ and cis-Ru(tpy)(AN)2Cl+ undergo photolysis in water to produce the same product with λmax= 475 nm. The photoproduct was characterized by electronic absorption spectroscopy and NMR as transRu(tpy)(H2O)2(AN)2+. As mentioned earlier, the acetonitrile ligand trans to the tpy ligand does not photolyze off. During the synthesis of cis-Ru(tpy)(AN)2Cl+ and trans-Ru(tpy)(AN)Cl2, no trace of the isomers trans-Ru(tpy)(AN)2Cl+ or cisRu(tpy)(AN)Cl2 could be detected, indicating these species do not generally form. The photolysis product was not Ru(tpy)(H2O)32+, it is a known species with a λmax= 532 nm48, which does not match the absorption of the photoproduct
34
4.3.1 [Ru(tpy)(AN)3]2+ The changes to the absorption spectra during the photoaquation of Ru(tpy)(AN)32+ in water are shown Figure 4.5. Figure 4.6 shows the photolysis followed by 1H NMR spectroscopy in D2O. Although it is not clear in the absorption spectra, the 1H NMR spectra shows an intermediate species forming during the reaction, most likely [Ru(tpy)(AN)2(H2O)]2+.
Figure 4.5 Photolysis of [Ru(tpy)(AN)3]2+ in water. 150W lamp 395 nm low pass filter. In the 1H NMR spectra, the highest field aromatic peak at 9 ppm and the peak corresponding to the equatorial acetonitrile at 2.9 ppm are the most informative because they shift the most.
They both shift upfield by about 0.1
ppm early in the photolysis, corresponding to the formation of the intermediate, and then shift upfield again by an additional ~0.2 ppm and split into two peaks. In fact, every peak in the spectrum of [Ru(tpy)(AN)3]2+ splits into two peaks when 35
the photolysis is complete. Figure 4.7 compares the 1H NMR spectra of [Ru(tpy)(AN)3]2+ and the photoproduct in D2O. The integration of the aromatic peaks remains the same, indicating no reaction with the tpy ligand. The equatorial acetonitrile peaks of the photoproduct also integrate to 3 hydrogens. The peak corresponding to the axial acetonitrile ligands also shifts upfield, but only by a total of 0.1 ppm, and it does not split. The lack of splitting indicates that the axial acetonitrile is no longer coordinated to the metal complex, and its peak position of 2.1 ppm has been identified as free acetonitrile in D2O.49 The integration of the free acetonitrile peak is six hydrogens, indicating all of the axial acetonitriles were photolized off. The splitting of the product peaks suggests that there may be more than one photoproduct. The spectrum after photolysis for 12 hours is identical to that of 24 hours, showing that the reaction had gone to completion. An impurity of N(Bu)4+ is also present in the spectra. The photolysis reaction is shown in equation (3).
36
B N N
N
Ru
N
N
A
N
B
Figure 4.6 Photolysis of Ru(tpy)(AN)32+ in D2O 150W lamp. A 400 nm band pass filter was used for the first 15 hours, and then switched to a 395 nm low pass filter for an additional 24 hours. (A) equatorial acetonitrile, (b) axial acetonitrile, and (c) N(Bu)4Cl impurity.
37
Figure 4.7 Ru(tpy)(AN)32+ in D2O (top) and photoproduct after photolysis for 24 hours in D2O with 150W lamp and a 395 nm low pass filter (bottom). (A) equatorial acetonitrile, (b) axial acetonitrile, (c) N(Bu)4Cl impurity and (d) free acetonitrile.
N
N N H3CCN
NCCH3 Ru
hv, H2O
NCCH3
N H3CCN
N
N O2H
Ru
N
NCCH3
hv, H2O
N HO2
O2 H Ru
NCCH3
N
(3) 4.3.2 cis-[Ru(tpy)(AN)2Cl]+ Cis-[Ru(tpy)(AN)2Cl]+ undergoes photoaquation to the same photoproduct as [Ru(tpy)(AN)3]2+. The photolysis followed by UV-Vis spectroscopy is shown in Figure 4.8. The photolysis of Ru(tpy)(AN)2Cl+ in D2O was also followed by 1H 38
NMR and yielded the same photoproduct, as seen in Figure 4.9. The only difference is the integration of the free acetonitrile ligand corresponds to 3 hydrogen atoms, as there is only one axial acetonitrile ligand. Ru(tpy)(AN)2Cl+ underwent photoaquation at wavelengths as long as 590 nm with comparable speed to that of 400 nm. This approaches the ideal window of 700-900 nm. The photolysis reaction is shown in equation 4.
Figure 4.8 Photolysis of [Ru(tpy)(AN)2Cl]+ in water. 150W lamp 395 nm low pass filter
N
N N
NCCH3 Ru
Cl
NCCH3
hv, H2O
N
O2H Ru
Cl N
N
N
NCCH3
hv, H2O
N HO2
O2 H Ru
NCCH3
N
(4) 39
Figure 4.9 Comparison of photolysis products of Ru(tpy)(AN)32+ in D2O (top) and Ru(tpy)(AN)2Cl+ in D2O (bottom). (A) equatorial acetonitrile, (b) free acetonitrile, and (c) N(Bu)4Cl impurity. 4.4 Quantum Yields The Quantum yield of photoanation with chloride and photoaquation in water were measured for both Ru(tpy)(AN)3 and Ru(tpy)(AN)2Cl with λirr = 400 nm. The quantum yield was measure from the slope of the decrease in concentration of the reactant with irradiation time. Such a graph for the photoanation of Ru(tpy)(AN)3 is shown in Figure 4.10. The quantum yields of Ru(tpy)(AN)3 and Ru(tpy)(AN)2Cl are listed in Table 4.2. The quantum yield of photoanation with chloride was independent of the chloride concentration. An 40
excess of chloride ions with concentrations from ~5-15 mM was used.
Figure 4.10. Concentration of [Ru(tpy)(AN)3]2+ vs. irradiation time. Photolysis was done in CH2Cl2 in the presence of Cl-, 150W lamp 400nm band pass filter. Concentration of [Ru(tpy)(AN)3]2+= 250 µM. Concentration of chloride solution ~10 mM.
Complex Ru(tpy)(AN)32+
Quantum yield of photoanation with chloride in CH2Cl2 0.040 mol/einstein
Quantum yield of photoaquation 0.035 mol/Einstein
Ru(tpy)(AN)2Cl+
0.12 mol.einstein
0.12 mol/Einstein
Table 4.2 Quantum yields of photoaquation and photoanation. λirr = 400 nm.
The quantum yields of photoanation with chloride have been previously reported at λirr = 436 nm.40 The quantum yield of Ru(tpy)(AN)32+ is the same at both wavelengths. The quantum yield of Ru(tpy)(AN)2Cl+ is reported as 0.13 at λ41
irr
= 436. Ru(tpy)(AN)2Cl+ absorbs much less at 400 nm than it does at 434 nm.
Examination of the UV-Vis spectra shows a second peak as a shoulder to the large intensity ligand centered transition that tails into the 400 nm range. Irradiation at 400 nm could excite both of these states, and if the higher energy transition doesn’t lead to ligand loss this could explain the lower quantum yield. 4.5 Dark Reactions In order to study the photolysis of ruthenium terpyridine compounds, the reactivity in the absence of light had to be investigated to make sure there was no dark reaction effecting the measurement of the quantum yields. The dark reactions of [Ru(tpy)(AN)3]2+ and [Ru(tpy)(AN)2Cl]+ in water and in CH2Cl2 with TBACl were studied. The solutions were degassed by bubbling N2 through them for 10 min, and allowed to sit for at least 1 day in the dark. The UV-Vis spectrum of each dark reaction is shown in Figures 4.11 and 4.12. In every case the λmax of the compound is unchanged indicating that there was no reaction. In some of the spectra, the absorbance taken after one day has a higher intensity then the initial absorbance. This is because the samples were degassed immediately after the initial absorbance was taken, and a loss of solvent during the degassing process concentrates the sample. This is especially prevalent in for Ru(tpy)(AN)2Cl in CH2Cl2.
42
(a)
(b)
Figure 4.11 Dark reaction of (a) [Ru(tpy)(AN)3]2+ and (b) cis-[Ru(tpy)(AN)2Cl]+ in water. The solid line is the initial absorbance; the dashed line is after 25 hours. 43
(a)
(b)
Figure 4.12 Dark reaction of (a) [Ru(tpy)(AN)3]2+ and (b) cis-[Ru(tpy)(AN)2Cl in CH2Cl2 with TBACl. The solid line is the initial absorbance; the dashed line is after 25 hours.
44
CHAPTER 5
Ruthenium Bis-dipyridyl Benzonitrile Compounds
5.1 Electronic Absorption Spectra The electronic absorption spectrum of cis-[Ru(bpy)2(NCPh)2]2+ has been previously reported.52 The absorption spectra of the cis-[Ru(bpy)2L2]2+ ( L= benzonitrile,
NCPh,
4-fluorobenzonitrile,
NCPhF,
4-methylbenzonitrile,
NCPhMe, and 4-methoxybenzonitrile, NCPhOMe) compounds are shown in Figure 5.1. The absorption spectra show 1ππ* transitions centered on the bpy ligand at ~280 nm, and a 1MLCT from the ruthenium t2g state to the π* orbital of the bpy ligand with a maxima at ~410-420 nm. The absorption maximum and molar extinction coefficient for each compound is listed in Table 5.1
45
Figure 5.1 Absorption spectra of cis-[Ru(bpy)2(NCPh)2]2+ (solid line), cis[Ru(bpy)2(4-Me-NCPh)2]2+ (dashed line), cis-[4-OMe-Ru(bpy)2(NCPh)2]2+ (X’s) and cis-[Ru(bpy)2(4-F-NCPh)2]2+ (dotted line) in water.
Absorption λmax, nm ( ε, M-1cm-1 ) 2+ cis-[Ru(bpy)2(AN)2] 420 (8900)a 2+ cis-[Ru(bpy)2(4-F-NCPh)2] 412 (6500) cis-[Ru(bpy)2(NCPh)2]2+ 413 (6500) cis-[Ru(bpy)2(4-Me-NCPh)2]2+ 420 (6500) Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ 417 (8200) cis-[Ru(bpy)3]2+ 451 (14800)b Table 5.1 Absorption maximum and extinction coefficients of cis-[Ru(bpy)2L2]2+ compounds in water. A from reference 56. B from reference 57. Compound
The absorption maximum for the cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-MeNCPh, 4-OMe-NCPh, and 4-F-NCPh) range from 412 to 420 nm, which is similar to the absorption of cis-[Ru(bpy)2(AN)2]2+ at 420 nm. [Ru(bpy)3]2+ has a lower energy MLCT maximum of 451 nm. Among the benzonitrile ligands there is a trend with the electron donating methyl and methoxy substituents having lower energy absorption. A compound that has greater π-back bonding would have a higher
energy
MLCT
transition because of the lowering of the metal t2g 46
orbitals. Cis-[Ru(bpy)3]2+ has the lowest energy transition because it has much weaker back-bonding than the nitrile ligands. The benzonitrile ligands with the electron withdrawing groups have higher energy MLCT transitions because they have greater π-back bonding. Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ ligand does not fit the trend, having an absorption maxima at higher energy than cis-[Ru(bpy)2(4Me-NCPh)2]2+. 5.2 Photolysis in Water Irradiation of the 1MLCT absorption band of cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, and 4-F-NCPh) in water results in the substitution of the benzonitrile ligands by water. The final product of photolysis is the diaqua cis-[Ru(bpy)2(H2O)2]2+ compound. Cis-[Ru(bpy)2(H2O)2]2+ is a known species with an absorption maximum of 484 nm53 in water, and can be identified as the product of the photolysis of each compound. An intermediate is also obsearved for each compound with a maximum absorbance at an intermediate wavelength between the starting compound and the final product. The intermediate for each compound has a λmax~ 450 nm, and is presumed to be the monoaqua species [Ru(bpy)2(H2O)L]2+. The photolysis reaction is shown in equation 1.
N N
N
L Ru
N
N
N
L
hv, H2O
N
N
Ru
N
OH2
L
N hv, H2O
N
OH2 Ru
OH2
N
(1)
47
The reaction is analogous to the irradiation of cis-[Ru(bpy)2(AN)2] 2+ in water, which also shows a mono-aqua intermediate with a λmax= 458 nm, followed by the formation of the bis-aqua species.49 The photolysis of cis[Ru(bpy)2(AN)2]2+ in water is known to be a two photon process, requiring a second photon to form the bis-aqua species from the mono-aqua species49, and the photolysis of the cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, and 4-F-NCPh) series discussed here is most likely also a stepwise two photon process. 5.2.1 Cis-[Ru(bpy)2(NCPh)2]2+ and cis-[Ru(bpy)2(4-Me-NCPh)2]2+ Cis-[Ru(bpy)2(NCPh)2]2+ was irradiated at 450 nm in water and followed by electronic absorption spectroscopy. At early times an intermediate formed with λmax
=
444
nm
assigned
[Ru(bpy)2(NCPh)(H2O)]2+,
to
the
monosubstituted
species
cis-
before converting to the final product, cis-
[Ru(bpy)2(H2O)2]2+, with λmax = 490 nm. The changes to the absorption spectra as a function of photolysis time are shown in Figure 5.2. Cis-[Ru(bpy)2(4-Me-NCPh)2]2+ was irradiated at 450 nm in water and the changes in its electronic absorption spectra was followed. At early times cis[Ru(bpy)2(4-Me-NCPh)(H2O)]2+ with maximum at 447 nm was formed, before converting to the product, cis-[Ru(bpy)2(H2O)2]2+. The changes to the absorption spectra are shown in Figure 5.3.
48
(a)
(b)
Figure 5.2 Photolysis of cis-[Ru(bpy)2(NCPh)2]2+ in water(a) 0-1 minute of irradiation and (b) 1.5-20 minutes of irradiation with 150W lamp 450 band pass filter FWHM=70 nm. 49
(a)
(b)
Figure 5.3 Photolysis of cis-[Ru(bpy)2(4-Me-NCPh)2]2+ in water (a) 0-2.5 minutes of irradiation and (b) 3-20 minutes of irradiation with 150W lamp 450 band pass filter FWHM=70 nm. 50
5.2.2 Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ and cis-[Ru(bpy)2(4-F-NCPh)2]2+ Similar results to those described above were observed for cis[Ru(bpy)2L2]2+ (L = 4-OMe-NCPh, 4-F-NCPh) upon photolysis. For Cis[Ru(bpy)2(4-OMe-NCPh)2]2+, irradiation with 450 nm light in water results in the initial formation of cis-[Ru(bpy)2(4-OMe-NCPh)(H2O)]2+ with maxima at 448 nm,
before converting to the product, cis-[Ru(bpy)2(H2O)2]2+ (Figure 5.4.)
Irradiation of cis-[Ru(bpy)2(4-F-NCPh)2]2+ at 450 nm in water also shows the formation of the mono-aqua intermediate, cis-[Ru(bpy)2(4-F-NCPh)(H2O)]2+ (λmax = 441 nm) prior to the generation of cis-[Ru(bpy)2(H2O)2]2+ (λmax = 490 nm), as shown in Figure 5.5.
51
(a)
(b)
Figure 5.4 Photolysis of cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ in water (a) 0-4 minute of irradiation and (b) 4-20 minutes of irradiation with 150W lamp 450 band pass filter FWHM=70 nm. 52
(a)
(b)
Figure 5.5 Photolysis of cis-[Ru(bpy)2(4-F-NCPh)2]2+ in water (a) 0-2 minute of irradiation and (b) 2-18 minutes of irradiation with 150W lamp 450 band pass filter FWHM=70 nm. 53
5.2.4 Rates of Photolysis In order to test how changing the ligand affects the rate of the photolysis of the cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, and 4-F-NCPh) series, each compound was irradiated with a 400 nm band pass filter (FWHM = 10 nm) and a graph was made of the absorbance vs. time. The slopes of the graphs at early times were compared as a measure of the speed of photolysis. The 400 nm band pass filter was chosen to irradiate the starting material with minimal irradiation of the intermediate and product in order to avoid secondary photolysis reactions stemming from those species. Only the rate of the first step of the photolysis was measured, producing the mono-aqua species. The change in absorbance was measured as the increase at 450 nm, corresponding to the intermediate mono-aqua species, for the first 5 minutes of irradiation. None of the bis-aqua species could be seen within the first 5 minutes with the 400 nm band pass filter. In order to normalize the rates of photolysis of each compound to each other, the change in absorbance with time was divided by the intensity of the lamp, measured by ferricoxalate actinometry, according to equation 2.
In
equation 2, ∆A is the change in absorption at 450 nm, ∆t is the irradiation time, I is the intensity of the lamp, and ρ represents the rate of photolysis. The results of the application of equation 2 for the photolysis of each compound are listed in Table 5.2.
54
∆A ρ = ∆t I
(2)
Rate of photoaquation, ρ Relative quantum yield, Φ (absorbance/time*I) (mol/einstein) AN 970 0.21 NCPhF 940 0.20 NCPh 720 0.16 NCPhMe 390 0.08 NCPhOMe 620 0.13 Table 5.2 Rates of photolysis in water of the cis-[Ru(bpy)2L2]2+ series (λirr = 400 nm). L
Complexes of the type cis-[Ru(bpy)2L2]2+ with L = acetonitrile have larger quantum yields of ligand loss than complexes with L = NH3 or Py. It is hypothesized that the large quantum yield of acetonitrile is due to π-back bonding with the metal. Excitation of the 1MLCT disrupts the π-back bonding by removing electron density from the metal, causing ligand loss. The compounds cis[Ru(bpy)L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, and 4-F-NCPh) was synthesized to test this hypothesis. By varying the functional group at the para position of the coordinated benzonitrile, the electron density on the ligand can be changed. Adding electron withdrawing groups should increase the back bonding character of the metal ligand bond, causing an 1MLCT transition to have a larger effect on the strength of the ruthenium benzonitrile bond. This would correlate as an increase in photoinduced ligand loss. An electron donating group should reduce the back bonding character of the metal ligand bond, causing the 55
photoinduced ligand loss to decrease. Three of the compounds, cis[Ru(bpy)2(NCPh)2]2+, cis-[Ru(bpy)2(4-Me-NCPh)2]2+, and cis-[Ru(bpy)2(4-FNCPh)2]2+ followed the expected trend. The electron withdrawing fluoride group sped up the photolysis of cis-[Ru(bpy)2(4-F-NCPh)2]2+, while the electron donating methyl group slowed down the photolysis of cis-[Ru(bpy)2(4-MeNCPh)2]2+, and cis-[Ru(bpy)2(NCPh)2]2+ photolyzed at an intermediate speed. Cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ was expected to have the slowest photolysis due to the electron donating methoxy group, but instead it showed a photolysis speed between that of cis-[Ru(bpy)2(4-Me-NCPh)2]2+ and cis-[Ru(bpy)2(NCPh)2]2+. The rate of photolysis for each compound was compared to the rate of photolysis of cis-[Ru(bpy)2(AN)2]2+, for which a quantum yield of 0.21 is known.49 From this a relative quantum yield for each other compound could be calculated, and is recorded in Table 5.2 with the rates of photolysis. The electronic absorption maximum (Table 5.1) is also a measure of the πback bonding in the complex. Increased π-back bonding causes the metal t2g orbitals to decrease in energy, which causes an increase in the energy of the MLCT absorption from the t2g orbital on the metal to the bpy ligand. The electronic absorption maximum and the photolysis rates follow the same trend with
cis-[Ru(bpy)2(4-OMe-NCPh)2]2+
having
an
intermediate
electronic
absorption and photolysis rate between cis-[Ru(bpy)2(NCPh)2]2+
and cis-
[Ru(bpy)2(4-Me-NCPhMe)2]2+. Although cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ has the methoxy group on the benzonitrile ring which should be more electron donating than a methyl group, it appears that it has more π-back bonding 56
than cis-[Ru(bpy)2(4-Me-NCPh)2]2+. A graph of electronic absorption maxima vs. photolysis rate for the series cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMeNCPh, 4-F-NCPh) is shown in Figure 5.6. A trend can be seen showing increased photolysis rate with increased energy of absorption, indicating that it is the π-back bonding that affects the photolysis rate.
Figure 5.6 Absorption maxima in water vs. rate of photolysis in water of the series of compounds cis-[Ru(bpy)2L2]2+ (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, 4-F-NCPh).
5.3 Electrochemistry The electrochemical behavior of the cis-[Ru(bpy)2L2]2+ compounds (L = NCPh, 4-Me-NCPh, 4-OMe-NCPh, 4-F-NCPh, bpy, and acetonitrile) was studied in distilled acetonitrile with 0.1 M N(Bu)4(PF6) as the supporting electrolyte. The electrochemical potentials are shown in Table 5.3. 57
Compound E1/2 ([Ru]3+/2+)a/ E1/2 ([Ru]2+/1+)a / E1/2 ([Ru]1+/0)a / Cis-[Ru(bpy)2L2]2+ V V V L = AN +1.70 -1.11 -1.30 L = 4-F-NCPh +1.72 -0.99 -1.17 L = NCPh +1.80 -1.08 -1.23 L= 4-Me-NCPh +1.81 -1.10 -1.28 L = 4-OMe-NCPh +1.70 -1.11 -1.22 2+b [Ru(bpy)3] +1.54 -1.07 -1.26 Table 5.3 Half-wave potentials E1/2 (V) vs NHE of complexes [Ru(bpy)2L]2+ (L = AN, 4-F-NCPh, NCPh, 4-Me-NCPh, 4-OMe-NCPh, and bpy). (A) [Ru] represents the complex. (B) from reference 58.
The cyclic voltammograms of these compounds are consistent with metalbased reversible oxidations and several ligand-based reductions.59 The reduction potentials for the cis-[Ru(bpy)2L2]2+ series is similar to that of [Ru(bpy)3]2+, indicating that the two reduction potentials are to different bpy ligands. The oxidation potentials show that the cis-[Ru(bpy)2L2]2+ series of complexes are harder to oxidize than [Ru(bpy)3]2+. This is due to the stabilization of the t2g orbitals by π-back bonding afforded by the benzonitrile ligands. The similarity in the oxidation potential of the cis-[Ru(bpy)2L2]2+ series shows there is little variation in the π-back bonding strength of the different ligands, however it appears that NCCH3, NCPhF, and NCPhOMe have more π-backbonding than NCPhMe and NCPh. The different in photolysis rates can than be explained by the strength of the sigma bonding of the ligands, the stronger the sigma bonding the slower the rate of photolysis. The compound with L = 4-OMe-NCPh doesn’t fit the trend. It is expected to have the most sigma bonding due to its electron donor ability, but does not show the slowest photolysis rate, nor the largest oxidation potential. 58
5.4 Dark Reactions In order to understand the effect of light on the reaction cis-[Ru(bpy)2L2]2+ (L = 4-F-NCPh, NCPh, 4-Me-NCPh, and 4-OMe-NCPh) compounds, the reactivity in the absence of light was investigated. Each compound was dissolved in water and kept in the dark for 24 hours. The spectra showing the initial absorbance along with the absorbance after 1 day are shown in Figures 5.7 and 5.8. The spectra of cis-[Ru(bpy)2(NCPh)2]2+ and cis-[Ru(bpy)2(40-OMeNCPh)2]2+ show no change over 24 hours. Cis-[Ru(bpy)2(4-F-NCPh)2]2+ shows a trace amount of conversion to the bis-aqua species, indicating a dark reaction but at a rate that is negligible compared to the time scale of the photolysis (10 min). The spectrum of cis-[Ru(bpy)2(4-Me-NCPh)2]2+ shows a small change after 24 hours to the bis-aqua species. However, there is negligible change after 2 hours, demonstrating that the dark reaction does not affect the photolysis reactions which are complete within 30 minutes. It was hypothesized for the series cis-[Ru(bpy)2L2]2+ (L = 4-F-NCPh, NCPh, 4-Me-NCPh, and 4-OMe-NCPh) that the more electron withdrawing ligands would have more π-back bonding with the metal. This increased π-back bonding is expected to increase the rate ligand loss upon excitation of the MLCT transition. Three of the compounds (L = 4-Me-NCPh, NCPh, and 4-F-NCPh) follow the expected trend. However, the electrochemical data shows similar oxidation potentials for all three complexes and identical oxidation potentials for L = NCPh and 4-Me-NCPh. This suggests that it is not only the amount of π-back bonding, but also the amount of σ
bonding that affects the rate of photolysis. 59
The complex with L = 4-OMe-NCPh doesn’t fit the trend.
(a)
(b)
Figure 5.7 Dark reaction of (a) cis-[Ru(bpy)2(NCPh)2]2+ and (b) cis-[Ru(bpy)2(4Me-NCPh)2]2+ in water. Solid line is intial absorbance, dotted line is after 24 hours, dashed line is after 2 hours (cis-[Ru(bpy)2(NCPhMe)2]2+ only). 60
(a)
(b)
Figure 5.8 Dark reaction of (a) cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ and (b) cis[Ru(bpy)2(4-F-NCPh)2]2+ in water. Solid line is intial absorbance, dotted line is after 24 hours.
61
CHAPTER 6
CONCLUSION
Metal complexes have been used to treat tumors for decades. Cisplatin and its derivatives are by far the most commonly used, and have been shown to be effective against several types of cancer.1,2 The major drawback of chemotherapy with cisplatin and its analogs is the side effects associated with its use. Cisplatin is not very selective towards tumor tissue and can cause significant damage to healthy cells.9 Acquired resistance to cisplatin over time is also a problem.10,12 Ruthenium compounds have been studied as alternatives in an attempt to alleviate these problems. Several ruthenium compounds including Ru(azp)2Cl2 and cisRu(DMSO)4Cl2 have been shown to bind to DNA and kill cells by thermal substitution to the aqua species.38,39 One way to increase selectivity is to select a compound that doesn’t thermally bind DNA, but becomes active upon irradiation with light. Using light in the treatment of tumors is known as photodynamic therapy (PDT). Since mammalian cells have a high absorbance in the UV range, an ideal absorbance for a potential PDT drug is in the 700-900 nm range for maximum tissue penetration. A high quantum yield of photosubstitution is also desirable. Ruthenium compounds bound to imine ligands such as bpy and tpy 62
have been shown to undergo photoaquation both thermally and upon irradiation.21,31,32,37-41 Ruthenium compounds bound to tpy ligands exhibit a lower energy MLCT transition compared to those with coordinated
bpy ligands (Figure 2.4). In
conjunction with acetonitrile ligands, these complexes exhibit large quantum yield of
photosubstitution.
For
these
reasons
[Ru(tpy)(AN)3]2+
and
cis-
[Ru(tpy)(AN)2Cl]+ were chosen and studied as possible candidates for PDT agents. Both compounds were shown to photolyze to the diaqua product trans[Ru(tpy)(AN)(H2O)2]2+ upon irradiation with visible light. Cis-[Ru(tpy)(AN)2Cl]+ has a lower energy MLCT, and photolysis can be achieved with irradiation wavelengths as low as 590 nm.
It exhibits a greater quantum yield of
photosubstitution in water (0.12) compared to 0.035 for Ru(tpy)(AN)32+ under similar irradiation conditions. As such Ru(tpy)(AN)2Cl+ is the most likely candidate for PDT. The series of compounds cis-[Ru(bpy)2L2]2+ (L = 4-F-NCPh, NCPh, 4Me-NCPh, and 4-OMe-NCPh) shown in Figure 2.5 were studied to gain further understanding for the high quantum yield of photosubstitution of acetonitrile compared to other monodentate ligands bound to a ruthenium(II) center. It is suspected that π-back bonding to the ruthenium atom by CH3CN is disrupted upon MLCT excitation. By varying the substituents in the para position of the benzonitrile ligands, the electron density of the ligand can be changed. Electron withdrawing groups should increase the π-back bonding to the metal, while electron donating groups should
decrease it. It was predicted that an 63
increase in π-back bonding should correlate with an increase in quantum yield of photosubstitution, while a decrease in π-back bonding should have the reverse effect. The rate of photolysis of each compound was measured in water. Three of the compounds, cis-[Ru(bpy)2L2]2+ (L =
4-FNCPh, NCPh, and 4-MeNCPh)
follow the trend of decreasing rate of photolysis with electron donating substituents. However, cis-[Ru(bpy)2(4-OMe-NCPh)2]2+ exhibits a photolysis rate between that of the benzonitrile complex and the 4-methylbenzonitrile complex (Table 5.2).
The electrochemical data shows more π-back bonding in cis-
[Ru(bpy)2(NCPh)2
and
cis-Ru(bpy)2(4-Me-NCPh)2]2+
Ru(bpy)2(4-OMe-NCPh)2]2+,
compared
cis-Ru(bpy)2(4-F-NCPh)2]2+,
and
to
ciscis-
Ru(bpy)2(AN)2]2+. The absorption maxima of the MLCT transition is also a measure of the π-back bonding, since the metal-centered HOMO is stabilized while the reduction of the ancillary bpy ligands is relatively unaffected. The rate of photolysis of the series tracks well with absorption maximum, indicating that π-back bonding is important in the observed photoreactivity. The original hypothesis is correct, the more electron withdrawing substituents increase the rate of photosubstitution of benzonitrile ligands, and therefore are better candidates for PDT therapy.
64
REFERENCES (1) Sherman, S. E.; Lippard, S. J. Chem. Rev. 1987, 87, 1153-1181 (2) Loehrer, P. J.; Einhorn, L. H. Annal. Int. Med. 1984, 100, 704-713 (3) Fichtinger-Schepman, A. J.; Van der Veer, J. L.; Den Hartog, J. H.; Lohman, P. H.; Reedijk, J. Biochemistry. 1985, 24 (3) 707-713 (4) Eastman, A. Biochemistry, 1986, 25 (13), 3912-3915 (5) Hambley, T. W.; Jones, A. R. Coordination Chemistry Reviews 2001, 212, 35-59 (6) Gelasco, A.; Lippard, S. J. Biochemistry, 1998, 37 (26), 9230-9239 (7) Huang, J.; Zamble, D.; Reardon, J.; Lippard, S. J.; Sancar, A. Proc. Natl. Acad. Sci. USA 1994, 91, 10394-10398 (8) Bruhn, S. L.; Pil, P. M.; Essigmann, J. M.; Housman, D. E.; Lippard S. J. Proc. Natl. Acad. Sci. USA, 1992, 89, 2307-2311 (9) Pil, P. M.; Lippard, S. J. Science, 1992, 256, 234-237 (10) Chu, G. J. Biol. Chem. 1994, 269, 787-790 (11) Sanderson, B. J.; Ferguson, L. R.; Denny, W. A. Mutation Research 1996, 355, 59-70 (12) Kartalou, M.; Essigmann, J. M. Mutation Research 2001, 478, 23-43 (13) Pushpan, S. K.; Venkatraman, S.; Anand, V. G.; Sankar, J.; Parmeswaran, D.; Ganesan, S.; Chandrashekr, T. K. Curr. Med. Chem.-Anti-Cancer Agents, 2002, 2, 187-207 (14) Lutterman, D. A.; Fu, P. K.; Turro, C.; J. Amer. Chem. Soc. 2006, 128, 738-739 (15) Sorasaenee, K.; Fu, P. K.; Angeles-Boza, A. M.; Dunbar, K. R.; Turro, C. Inorg. Chem. 2003, 42, 1267-1271 (16) Turro, C.; Hall, D. B.; Chen, W.; Zuilhof, H.; Barton, J. K.; Turro, N. J. J. Phys. Chem. A. 1998, 102, 5708-5715
65
(17) Angeles-Boza, A. M.; Bradley, P. M.; Fu, P. K.; Shatruk, M.; Hilfiger, M. G.; Dunbar, K. R.; Turro, C. Inorg. Chem. 2005, 44 7262-7264 (18) Chifotides, H. T.; Hess, J. S.; Angeles-Boza, A. M.; Galan-Mascaros, J. R.; Sorasaenee, K.; Dunbar, K. R. Daltan Trans. 2003, 4426-4430 (19) Angelas-Boza, A. M.; Bradley, P. M.; Fu, P. K.; Wicke, S. E.; Bacsa, J.; Dunbar, K. R.; Turro, C. Inorg. Chem. 2004, 43, 8510-8519 (20) Chifotides, H. T.; Koshlap, K. M.; Prez, L. M.; Dunbar, K. R. J. Am. Chem. Soc. 2003, 125, 10714-10724 (21) Chifotides, H. T.; Dunbar, K. R.; Chem. Eur. J. 2008, 14, 9902-9913 (22) Kang, M.; Chifotides, H. T.; Dunbar, K. R.; Biochemistry, 2008, 47, 2265-2276 (23) Chifotides, H. T.; Dunbar, K. R.; Chem. Eur. J. 2006, 12, 6458-6468 (24) Chifotides, H. T.; Koshlap, K. M.; Prez. L. M.; Dunbar, K. R. J. Am. Chem. Soc. 2003, 125, 10703-10713 (25) Fu, P. K.; Bradley, P. M.; Turro, C. Inorg. Chem. 2003, 42, 878-884 (26) Singh, T. N.; Turro, C.; Inorg. Chem. 2004, 43, 7260-7262 (27) Yao, L.; Hammitt, R.; Lutterman, D. A.; Joyce, L. E.; Thummel, R. P.; Turro, C. Inorg. Chem. 2009, 48, 375-385 (28) Bacac, M.; Hotze, A. C.; Van der Schilden, K.; Haasnoot, J. G.; Pacor, S.; Alessio, E.; Sava, G.; Reedijk, J. J. Inorg. Biochem. 2004, 98, 402-412 (29) Karidi, K.; Garoufis, A.; Hadjiliadis, N.; Lutz, M.; Spek, A. L.; Reedijk, J. Inorg. Chem. 2006, 45, 10282-10292 (30) Corral, E.; Hotze, A. C.; Magistrato, A.; Reedijk, J. Inorg. Chem. 2007, 46, 6715-6722 (31) Hotze, A. C.; Broekhuisen, M. E.; Velders, A. H.; Van der Schilden, K.; Haasnoot, J. G.; Reedijk, J. Eur. J. Inorg. Chem. 2002, 369-376 (32) Hotze, A. C.; Van der Geer, E. P.; Caspers, S. E.; Kooijman, H.; Spek, A. L.; Haasnoot, J. G.; Reedijk, J. Inorg Chem. 2004, 43, 4935-4943 (33) Chouai, A.; Wicke, S. E.; Turro, C.; Bacsa, J.; Dunbar, K. R.; Wang, D.; Thummel, R. P. Inorg. Chem. 2005, 44, 5996-6003
66
(34) Fu, P. K.; Bradley, P. M.; Van Loyen, D.; Drr, H.; Bossmann, S. H.; Turro, C. Inorg. Chem. 2002, 41, 3808-3810 (35) Tanya N. Singh. Ru(II) Complexes as Photoactivated Cisplatin Analogs. Ph.D. Dissertation, The Ohio State University, Columbus, OH, 2006 (36) Pinnick. D. V.; Durham, B. Inorg. Chem. 1984, 23, 1440-1445 (37) Clarke, M. J.; Coord. Chem. Rev. 2002, 232, 69-93 (38) Velders, A. H.; Kooijman, H.; Spek, A. L.; Haasnoot, J. G. de Vos, D.; Reedijk, J. Inorg. Chem. 2000, 39, 2966-2967 (39) Hotze, A. C.; Velders, A. H.; Ugozzoli, F.; Biagini-Cingi, M.; Manotti-Lanfredi, A. M.; Haasnoot, J. G.; Reedijk, J. Inorg. Chem. 2000, 39, 3838-3844 (40) Grover, N.; Gupta, N.; Thorp, H. H. J. Am Chem. Soc. 1992, 114, 3390-3393 (41) Novakova, O.; Kasparkova, J.; Vrana, O.; van Vliet, P. M.; Reedijk, J.; Brabec, V. Biochemistry 1995, 34, 12369-12378 (42) Cini, R.; Tamasi, G.; Defazio, S.; Corsini, M.; Zanello, P.; Messori, L.; Marcon, G.; Piccioli, F.; Orioli, P. Inorg. Chem. 2003, 42, 8038-8052 (43) Sullivan, P. B.; Calvert, J. M.; Meyer, T. J. Inorg. Chem. 1980, 19, 1404-1407 (44) Suen, H. F.; Wilson, S. W.; Pomerantz, M.; Walsh, J. L. Inorg. Chem. 1989, 28, 786-791 (45) Montalti, Marco; Credi, Alberto; Prodi, Luca; Gandolfi, Teresa M. Chemical Actinometry. Handbook of Photochemistry, third edition; Taylor and Francis: Boca Raton, FL, 2006; 603. ( 46) Ford, P.C. Coordination Chemistry Reviews, 44 1982 61-82 (47) Koike, K.; Okoshi, N.; Hori, H.; Takeuchi, K.; Ishitani, O.; Tsubaki, H.; Clark, I. P.; George, M. W.; Johnson, F. P. Turner, J. J. J. Am. Chem. Soc. 2002, 124, 11448-11455 (48) Adeyemi, A.; Dovletoglou, A.; Guadalupe, A. R.; Meyer, T. J. Inorg. Chem. 1992, 31, 13751383 (49) Yao, L.; Turner, D. B,; Singh, T. N.; Angeles-Boza, A. M.; Chouai, A.; Dunbar, K.R.; Turro, C. J. Am. Chem. Soc., 2009, 131, 26-27 (50) Assefa, Z.; Stanbury, D. J. Am. Chem. Soc. 1997, 119, 521-530 (51) Keene, F. R.; Salmon, D. J.; Meyer, T.
J. J. Am. Chem. Soc., 1976, 98, 1884-1889
67
(52) Adcock, P. A.; Keene, F. R.; Smythe, R. S.; Snow, M. R. Inorg. Chem., 1984, 23, 2336-2343 (53) Zhang, H.; Rajesh, C. S.; Dutta, P. K. J. Phys. Chem. A 2008, 112, 808-817 (54) Griffith, W. P.; Reddy, B.; Shoair, A. G.; Suriaatmaja, M.; White, A. J.; Williams, D. J. J. Chem. Soc. Dalton Trans., 1998, 2819-2825 (55) Bhuiyan, A. A.; Kincaid, J. R. Inorg. Chem. 2001, 40, 4464-4471 (56) Zhou, M.; Robertson, G. P.; Roovers, J. Inorg. Chem. 2005, 44, 8317-8325 (57) Ershov, A. Y.; Shashko, A. D.; Sizova, O. V.; Inanova, N. V. Russ. J. Gen. Chem. 2002, 72, 1413-1425 (58) Yugie Sun. Molecular Factors That Influence DNA Light Switch Behavior in Ruthenium(II) Complexes. M.S. Dissertation, The Ohio State University, Columbus, OH, 2007 (59) Liu, Y.; Chouai, A.; Degtyareva, N. N.; Lutterman, D. A.; Dunbar, K. R.; Turro, C. J. Am. Chem. Soc. 2005, 127, 10796.
68
