Design and applications of highly luminescent transition metal

technology and their applications. Particularly promising are devices based on the use of luminescence ma- terials as molecular reporters. Lumi- nesce...
1 downloads 0 Views 8MB Size
Download PDF
Design and Applications of Highly Luminescent I ransition Metal Complexes

H

-

d

e .

J. N. Demas Department of Chemistry University of Virginia Charlottesville, VA 22901

B. A. DeGraff

rently of interest because of their high sensitivity and specificity. In particular, remote luminescence sensors in fiber optic based systems are used for measuring oxygen, pH, pCO,, temperature, and for immu-

Department of Chemistry James Madison University Harrisonburg,VA 22807

There has been an explosion of inter est in the areas of probe and sensor technology and their applications. Particularly promising are devices based on the use of luminescence materials as molecular reporters. Luminescence probes and sensors are cur0003-2700/9 l /0363-829A/$02. 50/0 0 1991 American Chemical Society

noassays. Response is monitored by changes in luminescence intensity, lifetime (z), or spectral distribution (1-4). An increasingly important class of sensor materials is luminescent transition metal complexes, especially

those with platinum metals (Ru, Os, Re, Rh, and Ir). These materials have very desirable features. They can have long lifetimes (hundreds of nanoseconds to tens of microseconds), which makes their lifetimes much simpler and less expensive to measure than those of most organic fluorophores, which have low nanosecond lifetimes. Their luminescence quantum yields are independent of the excitation wavelength and can exceed 0.5, although values of 0.04 to 0.2 are more typical (5, 6). However, even these modest values are adequate for a variety of sensor applications. These complexes can show intense visible absorptions; this increases

ANALYTICAL CHEMISTRYl VOL. 63, NO. 17, SEPTEMBER 1,1991

829 A

REPORT sensitivity, simplifies sensor design, and expands the variety of excitation sources available. They also tend to be thermally, chemically, and photochemically robust, which aids sterilization and extends sensor lifetime. Figure 1 shows a typical absorption spectrum (Ru(bpy)$+,where bpy represents 2,2'-bipyridine) (7).Note the intense blue UV absorptions and the wavelength independence of the luminescence efficiency. Molecular engineering permits systematic alteration of spectroscopic and chemical properties. This chemical flexibility allows the design of systems that respond to specific environmental variables, permits either ionic or covalent attachment to a support or reagent, and lets absorption and emission properties be tailored to available excitation sources and detectors. In this REPORT we will describe the principles of designing highly luminescent metal complexes with specifically tailored properties for use as probes of polarity, pH, oxygen concentration, and chirality. To those unfamiliar with inorganic spectroscopy and photophysics, the pattern of luminescence for inorganic complexes can appear quite random and illogical. Table I lists representative metal complexes categorized by luminescence efficiency. The complex Ru(bpy),(CN),(H+),,, is a special case that will be discussed later. As we will explain, the rational design of luminescent materials with desired properties can be achieved by application of a few basic chemical and spectroscopic concepts.

We will first describe briefly the basic electronic structure and spectroscopic considerations of transition metal complexes, focusing on systems with d6 (denoting six d electrons) electronic configurations because this is one of the most promising areas, although the concepts are general. Our systems are limited to those containing at least one a-diimine ligand such as 2,2'-bipyridine (bpy) or 1,lOp henanthroline (p hen), a1t hough many of our design rules and fundamental principles also apply to other classes of luminescent metal complexes. The pertinent structurefunction relationships are demonstrated by case studies.

States of inorganic complexes We begin by examining the correlations between the electronic structure and the emission and absorption spectroscopy of metal complexes. Transition metal complexes are characterized by partially filled d orbitals (8). To a considerable extent the ordering and occupancy of these orbitals determine emissive properties. Figure 2 shows an orbital and state diagram for a representative octahed6 metal complex, where M dral is the metal and X is a ligand that coordinates at one site. The octahedral crystal field of the ligands splits the five degenerate d orbitals by an amount A into a triply degenerate t level and a doubly degenerate e level. The splitting arises because the two e orbitals are directed toward the six ligands and the remaining t, orbitals point between the ligands. The energetic consequences of this arrangement are attributable to the electrostatic interactions between the filled ligand orbitals and electrons placed in the different d orbitals. Thus an electron placed in an e orbital is of higher energy than one placed in a t, orbital. The magnitude of the splitting is given by the crystal

field splitting, A, which is a particularly important parameter whose size is determined by the crystal field strength of the ligands and the central metal ion. The luminescence properties of the complex can be controlled by altering the ligand, geometry, and metal ion. The distribution of electrons between the t, and e levels is profoundly affected by A. If A is large &e., strong field), it is energetically more favorable to pair electrons in the & level than to keep them unpaired by distributing them throughout the t, and e levels (Hund's rule). We will consider only strong crystal field systems where all six d electrons pair and fill the three tz orbitals (Figure 2). The ligands have n; and (T orbitals, but only the 'TI orbitals are spectroscopically important for visible- and near -UV absorptions and emissions. There are both 'TI bonding and 'TI antibonding ('TI*)levels, and the 'TI bonding levels are filled. The spectroscopic states are derived from the various orbital configurations (Figure 2). Because all spins are paired, the ground state is a singlet. The lowest excited states are derived from promoting a n electron to one of the unoccupied orbitals. The state classification is determined by the original and final orbitals. There are three types of excited states: metal-centered d-d states, ligandbased 'TI-'TI* s t a t e s , a n d chargetransfer states. We adopt the common usage that a d-d excited state is one derived by promoting a d electron to another d level; a 'TI-'TI* excited state is achieved by promoting a n electron from a 'TI bonding orbital to a 'TI* antibonding orbital; and a charge - transfer state is obtained either by promoting a d electron to a 'TI* antibonding orbital or promoting an electron in a 'TI bonding orbital to an unfilled d orbital. The lowest ligand excited states are 'TI-'TI* states derived from promot-

Table 1. Luminescence properties of d6 metal complexes Luminescent

Figure 1. Relative luminescence quantum yield (a) and absorption spectrum (b) of Ru(bpy)T in methanol at room temperature. The fall-off in luminescencequantum yields at the extremes are experimental artifacts. (Adapted from Reference 7.)

830 A

ANALYTICAL CHEMISTRY, VOL. 63, NO. 17, SEPTEMBER 1, 1991

Somewhat luminescent

Nonluminescent

ing a bonding n electron to a n* level. There are triplets and singlets, and the triplet is always below its analogous singlet state. These transitions are largely localized on the organic ligands and are spectroscopically very similar to those of t h e free ligand. The intense high-energy ab sorptions (250-300 nm) in Ru(bpy)g+ (Figure 1) are spin-allowed ligandlocalized n-n* transitions and are very much like those of the free bpy ligand. Similarly, singlet and triplet d-d states arise from promoting a bonding t electron to an e level (tgel). Transitions to d-d states are formally forbidden, even for t h e spinallowed singlet-singlet ones. Thus d-d emissions are characterized by long radiative lifetimes and high sus ceptibility to environmental quench ing. This results in very low or negligible luminescence yields, especially at room temperature. A further prob lem with exciting low-lying d-d states is that the t, orbitals are bonding with respect to the ligands while the e orbitals tend to be antibonding. Thus d-d states o f t ;e1 configuration tend to be very reactive with respect to decomposition by ligand displace ment. For these reasons we a r e aware of no d-d-emitting d6 molecules that are likely sensors. There are no visible d-d transitions i n Ru(bpy)F because they are too weak to be seen in the presence of the more intense charge- transfer and n-n* transitions. Charge -transfer (CT) states in volve both the organic ligand and the

metal. Metal - to - ligand charge trans fer (MLCT) involves promoting a n electron from a metal orbital to a ligand orbital (t;n*' configuration), and ligand-to-metal charge transfer (LMCT) involves promoting an electron from a ligand to a metal orbital (n'e'). a-Diimine ligands are easily reduced and, thus, the systems discussed here involve only MLCT transitions. CT transitions tend t o be more strongly allowed t h a n d-d transitions. Therefore they have shorter radiative lifetimes and are less susceptible to intramolecular and environmental quenching. Also, the molar extinction coefficients of t h e spin-allowed transitions a r e large, which makes these CT states easier to pump optically. The intense 454-nm visible band of Ru(bpy)g+ is an MLCT transition. None of our systems matches the ideal octahedral case of Figure 2. All have lower symmetry, which further splits the d levels. However, the model is a reasonable platform for further discussion because the types of states involved remain unchanged. The d state energies are still dictated by the average A of the ligands.

Design considerations An understanding of the above parameters allows synthetic control of the spectroscopy and, thus, the luminescence properties by alteration of the ligand, geometry, and metal ion. The most important design rule of luminescent transition metal complexes is that the emission always arises from the lowest excited state.

Figure 2. Simplified orbital and state diagrams for a d6 metal in an octahedral environment showing d and 71: bonding and n* antibonding orbitals. Each arrow represents an electron with its associated spin. A strong crystal field is assumed so that the t2 levels are filled. Ligand-to-metalcharge-transfer states are ignored.

This rule derives from the general principle that radiationless decay from upper excited levels is very fast, and relaxation to the lowest excited level occurs with nearly 100% efficiency. This is shown for Ru(bpy)p in Figure 1 where the luminescence efficiency is independent of excitation into either MLCT or n-x* excited states. The absence of fluorescences i n inorganic systems is a consequence of spin-orbit coupling from the metal, which accelerates intersystem crossing to t h e triplet manifold. This means that all major emission contributions in our systems will arise formally from the lowest triplet state, and the emissions will be forbidden phosphorescences. Thus control of the luminescence properties of complexes hinges on control of the relative state energies and the nature and energy of the lowest excited state. Our goal is the rational design of molecules that emit efficiently, have long lifetimes, are easily pumped, have specific environmental sensitivity, and are chemically and photochemically stable. Unfortunately, some of these aims are mutually antagonistic. Based on a large body of experimental and theoretical work, the following important rules have been found to dictate the photophysical and photochemical properties of our inorganic complexes: The lowest excited state must be either a CT or ligand n-n*. This prevent s photochemical instability . Any d-d states must be well above the emitting level to prevent their thermal excitation, which would result in photochemical instability and rapid excited-state decay. Spin-orbit coupling should be high, in order to increase the allowedness of the emission and permit radiative decay to compete more effectively with radiationless decay. Pure n-n* phosphorescences tend to be too long lived for efficient emission. Either spin-orbit coupling or mixing with more- allowed CT states must be used to increase the allowedness of n-n* phosphorescences. The emitting state cannot be too low in energy. The energy gap law states that radiationless processes become more efficient as the emitting state approaches the ground state (9, 10). One of the more important criteria is to remove the lowest d-d state from competition with the emitting level. Controlling the energies of the d-d states is accomplished by varying either the ligands or the central metal ion to affect A. Stronger crystal

ANALYTICAL CHEMISTRY, VOL. 63, NO. 17, SEPTEMBER 1,1991

831 A

REPORT field strength ligands or metals raise d-d state energies, and crystal field strength increases in the series (8): C1< py ,Cli is much lower than for Rh(bpy)3,’, and the d-d state for the chloro complex is the lowest state in the molecule giving the structureless emission. The CT states are at too high an energy to emit, because Rh(II1) is very difficult to oxidize or reduce. Similarly, Ru(bpy),(CN), emits well a t room temperature whereas Ru(bpy),(CN)Cl+ does not, because of the much lower effective crystal field strength of C1 versus CN. Ligand field strength can have significant impact on photostability. Consider cis-Ru(bpy),(py);+, where py has a smaller A than bpy. The absorption spectrum of Ru(bpy),(py):+ is virtually identical to that of Ru(bpy)z’. At 77 K, Ru(bpy),(py);+ and Ru(bpy)z+ give very similar, beautifully structured emissions. However, at room temperature, not only is Ru(bpy),(py);+ not emissive, it is one of the most photo-unstable molecules that we have seen; a solution in a cuvette will undergo substantial photodegradation while be ing carried across a lighted room to t h e spectrometer. The pyridine ligands a r e labilized t o give Ru(bpy),(py)S2+ and Ru(bpy),(S);+ where S is the solvent. The 77 K results show that the lowest excited state is MLCT, but the room-temperature photochemistry shows that the d-d state is thermally accessible from the excited MLCT state and provides t h e decomposition pathway. Figure 4, which shows the energy levels of Ru(bpy1,”’ a n d Ru(bpy),(py);+, illustrates this more clearly. The lower gap separating the d-d state from the emitting MLCT allows very efficient depopulation/ decomposition via the d-d state for Ru(bpy),(py);+ but not for Ru(bpy):+, where the gap is much larger. Photochemistry and deactivation via d-d states can be a problem (6, 14-17).For example, Ru(bpy):+ is deactivated in part via an upper d-d state; however, because of the higher A, the gap to the d-d state is larger and deactivation is less efficient than for the py complex. There is some d-d photochemistry with Ru(bpy);+. Because it is more difficult to remove a bidentate ligand, the yield for permanent photochemistry is much smaller than for the pyridine complex; the bpy can anneal by reattaching itself a t the vacant site (14,15). Because the deactivating d-d state is thermally activated, there is an

Cambridge Isotope Labs your supplier for

appreciable increase in the deactivation r a t e a s t h e temperature is raised. Thus Ru(bpy)z+ shows a strong change in lifetime above room temperature, which limits its use at higher temperatures. The thermal activation model is shown schematically in Figure 5a whereas Figure 5b shows a typical plot of the emission lifetime versus temperature. The temperature dependence of lifetime is

112 ( T )= (k, + k,,)

~

l3C & l2C i n the 90’s announces:

+ k’ exp( - AE/kT)

(I) where k , and k,, are the radiative and nonradiative decay constants going directly from the emitting state to the ground state, AE is the energy gap between the d-d state and emitting level, and k’ is the Arrhenius preexponential factor for thermal activation of the d-d state. This model assumes that decay from the d-d state is much more rapid than return to the MLCT (i.e., k , >> k p 1 ) ; therefore no thermal equilibrium is established. However, over the limited temperature range encountered by most sensors, a thermal equilibrium model and Equation 1 are indistinguishable. Fits to T versus T data allow measurement of the gap between the emitting MLCT state and the d-d state. The solid line best fit in

amham and

ammQnB NOW available from the world’s largest facility for the separation of carbon isotopes

12C in ~g quantities at 99.95 to 99.99% enrichments

l3C gases, intermediates & a complete line of labeled compounds and biochemicals

********** Please call now for a quotation on labeled starting materials or a custom synthesis. *YY+YYY**Y

Please inquire about our D, N, & 0 labeled compounds. For Your Free 1991-1992Catalog Please Contact: Figure 5. Energy degradation of Ru(bPY>? (a) State diagram showing the loss of excitation energy via radiationless decay through the d-d state. (b) Temperature dependence of the lifetime in a micellar media. The solid line is the best fit using a thermally activated deactivation via the d-d state. (Adapted from Reference 16.)

CIL CAMBRIDGE ISOTOPE LAB0 RATORIES 20 Commerce Way, Woburn, Massachusetts 01801 800-322-1174 (Toll-free)

I

617-9380067 (In Mass)

617-932-9721 (Fa)

CIRCLE 20 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL.63, NO. 17, SEPTEMBER 1,1991

833 A

SAMPLE PREPARATION BOMBS FOR TRACE METAL ANALYSIS Available in a variety of sizes and styles, our teflon-lined Acid Digestion Bombs are designed for microwave and convection heating. Use them to treat inorganic samples in HF, HCl and other strong mineral acids.Or to digest organic samples in strong alkali or oxidizing acid at elevated temperatures and pressures.

Our Oxygen Combustion Bombs decompose any combustible solid or liquid quickly and completely. The sealed bomb contains and converts materials to a soluble form while allowing complete recovery of all elements. Now available in a new longer-lasting material for high halogencontaining substances. For complete details ask for our free Brochure #4700.

REPORT Figure 5b was used to determine M. Thus Equation 1represents a reasonable starting model for predicting the temperature dependence of sensor behavior. Its suitability for microheterogeneous media remains to be tested. Thus a desirable design goal is to make the d-d state as thermally inaccessible as possible from the emitting MLCT or x-x* state. This can be done by increasing A by using a metal ion of the same oxidation state with a higher atomic number. The highest possible atomic number may seem to be desirable for optimum performance, but additional factors are also important. For example, compared with Ru(bpy)z+, Os(bpy);+ has a much larger splitting between the d-d and MLCT states because of the higher A of Os relative to Ru and because of the easier oxidation of Os, which lowers the MLCT s t a t e (Figure 2). Thus Os(bpy);+ shows no photochemistry or d-d deactivation. Unfortunately, the MLCT state has dropped so low that direct radiationless decay (energy gap law) reduces the lifetime to about 50 ns versus 600 ns for Ru(bpy)g+. Chemical modifications can be used to tune the state energies and enhance properties (17). CO ligands greatly stabilize the t levels, resulting in both a n increased A and a higher energy MLCT transition. For example, the primary MLCT bands of Os(phen)g+ and Os(phen),Cl(CO)+ are at 430 and 365 nm, respectively. The emissions are similarly shifted from 710 to 646 nm. In keeping with the expectations of the energy gap law, the lifetimes of Os(phen)g+ and Os(phen),Cl(CO)+ are 74 and 234 ns, respectively (18),and the former is somewhat luminescent, whereas the latter is luminescent (see Table I). An interesting point arises with Re(1) complexes (e.g., fac-Re+L(CO),X where X is a halide, pyridine, or nitrile) where the x - ~ * and MLCT states can be very nearly isoenergetic. This near-degeneracy allows the lowest s t a t e to be r a t h e r easily switched back and forth between x-x* and MLCT states by choosing suitable choices of ligands and in some cases by simply altering the temperature (19). Even when the x-x* state is lowest in Re(1) complexes, the emissions can be reasonably efficient in marked contrast to Rh(bpy)g+. We attribute this to two points. First, Re has a higher atomic number than Rh, it enhances spin-orbit coupling, and it increases t h e allowedness of t h e phosphorescence. Second, because of the proximity of the MLCT and x-x*

BPARR C INSTRUMENT OMPANY

211 Fifty-ThirdStreet Moline, IL 61265

309-762-771 6 270226 FOX: 309-762-9453

Telex:

CIRCLE 110 ON READER SERVICE CARD

834 A

ANALYTICAL CHEMISTRY, VOL. 63, NO. 17, SEPTEMBER 1,1991

states, mixing of the two will occur. Thus the less-allowed x-x* state “steals” allowedness from the moreallowed MLCT state, which increases the x-x* state’s radiative rate constant and enhances its emission efficiency. This illustrates an important principle. State mixing is frequently desirable because, when properly done, the best attributes of both states are incorporated. In the case just discussed, the MLCT state confers environmental sensitivity and a more efficient radiative rate. Concurrently, the x-x* character confers a longer lifetime. Finally, photochemically unstable ligands must be used with care. Re(dpk)(CO),Cl (dpk represents 2,2’dipyridyl ketone) shows a benzophenone-like phosphorescence at 77 K, indicating that the n-x* excited state of the ketone in the complex is the lowest state of the complex. No luminescence is seen at room temperature, and even at 77 K the dpk triplet state is such a powerful hydrogen atom extractor that it removes protons from alcohol glasses, as shown by the formation of the intense blue color of the keto free radical. The Ru(bpy),(CN),(H+),,, system is also interesting. The protonated forms do not seem to emit directly, but t h e excited- s t a t e protonated complex deprotonates rapidly to give the excited complex *Ru(bpy),(CN),, which does emit. In strong acid the overall efficiency is much lower (20). For this reason, this complex spans the “somewhat luminescent’’ and “nonluminescent” columns in Table I. Probe/sensor design We turn now to the question of how to incorporate specific sensitivity into molecular probes. When adding extra features, care must be taken not to violate the basic criteria established above. F o r d e s i g n of e n v i r o n m e n t a l probes, molecules that exhibit the largest solvent sensitivity a r e the most desirable. In general, asymmetric MLCT emitters a r e t h e best choice. Greatest sensitivity is found with complexes having the largest permanent dipole moment change when going from the ground state to the excited state. This dipole change interacts strongly with the solvent dipoles and causes large solvent effects on state energies and spectra. This result is analogous to that seen in the well - documented organic bio probes (21). Molecules showing this effect are frequently good environmental reporters.

High

For example, Ru(bpy)i+ is a poor choice for a n environmental probe because the promoted electron can distribute itself in a roughly spherical fashion with no overall dipole change and therefore shows almost no solvent effect on its absorption or emission. On the other hand, cisRu(bpy),(CN), h a s a n enormous change in the dipole on MLCT excitation because the charge is moved toward the two bpy ligands. cis-Ru(bpy),(CN), shows enormous solvatochromism in both emission and absorption as well as a large solvent sensitivity to lifetime. Solvatochromism is probably a requirem e n t for a n y system t h a t will demonstrate large environmental sensitivity. Other special probe properties can be built in by suitable ligand modification. For example, 5,6-dihydroxy1 , l O - phenanthroline complexes of Ru(I1) show pH sensitivity. facRe(bpy)(CO>,NC(CH,),CH~ ( n = 0-17) can provide hydrophobic hooks to attach relatively polar complexes to hydrophobic media such as micelles or cyclodextrins. This can be used to anchor the probe to specific structural features of the target (19). Temperature sensitivity can be achieved by having two states of different orbital types within kT of each other. Altering populations by temperature will then alter the decay rates, emission intensities, and lifetimes. Thermal activation of a d-d state is one choice, although careful design to minimize permanent decomposition becomes a problem. Crosby and coworkers have reported extraordinarily large changes in lifetime versus temperature relationships for Ru(II)L, complexes (T < 40 K) with negligible changes in the quantum yields (22).They have suggested the use of these complexes as cryogenic thermometers. For room temperature, some of the Re(1) systems with close n-n* and MLCT states are obvious choices; however, temperature dependence studies are just being started. For quenching-based oxygen sensors the dominant design consider ations appear to be the highest lifetimes and quantum yields (23-26). For a given class of complexes the bimolecular quenching constants are relatively insensitive to structure. Ruthenium(I1) complexes h a v e proved to be the most successful to date. Figure 6 shows a Stern-Volmer quenching plot for Ru(4,7Ph,phen);+ in silicone rubber and a luminescence intensity versus time plot of the same material while being

breathed over. The sensitivity of the luminescence intensity to 0, concentration is quite evident. Note the lower 0, concentration (higher emission intensity) on the initial exhalation (greater exchange time in the lungs), irregularities of the first few breaths because of the higher CO, levels in the blood, and restoration of the equilibrium concentration. Although general guidance can be given for sensorkomplex design, considerable work is still required to fabricate a useful device. In many cases the support matrix for the complex will alter its characteristics, sometimes in surprising ways. For example, we recently reported detailed photophysics and photochem istry of the Ru(I1)-based 0, sensors. The curvature of the Stern-Volmer quenching plots arises from ground state heterogeneity with different sites having different lifetimes and susceptibilities to oxygen quenching. For three different complexes, a two-site model gave superb fits to the quenching curves and is thus a potentially useful equation for fundamental studies and calibration (26).

We also examined oxygen quenching of Ru(I1) complexes bound to a silica surface-a potentially useful sensor support (27). These systems

Performance DvcorTM -----

Masg - - -

Spectrometers

The Dycor Quadrupole Mass Spectrometer offers a dynamic range of 7 orders of magnitude along with a high resolution CRT, analog bar, and tabular display modes, with an RS-232 port for computer interface as standard features. The Dycor product line is manufactured at our facility in the USA. This permits us to offer it at a price which is the most cost-effective in the industry. Whether your need is residual gas analysis, process monitoring, or leak detection, the microprocessor-based models provide you with the ultimate in performance.

Applications include: Residual Gas Analysis Process Monitoring Leak Detection Chemical Vapor Deposition Fermentation Sputtering Plasma Etching Molecular Beam Epitaxy Cryogenics High Energy Physics Vacuum Furnaces Evaporation Ion Beam Milling

Features:

Figure 6. Quenching of luminescence of Ru(4,7-Ph2phen);+ in silicone rubber. (a) Stern-Volmer quenching plot. The solid line is the best fit for a two-site model with different quenching constants. (Adapted from Reference 26.) (b) Luminescence intensity while being breathed over. (Adapted from Reference 25.)

1-100 or 1-200 AMU Range Faraday Cup and Electron Multiplier 9" or 12" High Resolution CRT Analog Bar or Tabular Display Pressure vs. Time Display Linear to 4 Decade Log Scale RS-232 Computer Interface 10-14 Torr Minimum Detectable Partial Pressure Background Subtraction Spectral Library Sample Systems for higher Pressures For literature, contact AMETEK, Process and Analytical Instruments Div. 150 Freeport Road, Pittsburgh, PA 15238, TEL: 41 2-828-9040, FAX: 412-826-0399.

PROCESS& ANALYTICAL INSTRUMENTS DIVISION

CIRCLE 3 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL. 63, NO. 17, SEPTEMBER 1,1991

835 A'

REPORT show even greater ground - state het erogeneity than the silicone systems. The modeling indicates that quenching by surface-adsorbed oxygen is probably important, a n d a twoparameter fitting equation based on a Freundlich adsorption isotherm beautifully fits the data for three complexes. Design of long-lived probes is still a compelling goal. This will allow greater sensitivity, easier lifetime measurements, and a longer time

scale in rotational probes. Our more recent work shows that a series of Re(1) complexes have great promise as sensors and as molecular probes. Their unquenched lifetimes can exceed 100 ys with luminescence quantum yields of > 0.5. Trigonal ML, metal complexes exist as optically active pairs and can show enantiomeric selective binding to DNA as well as selective excitedstate quenching. One of the optically active enantiomers of RuL, complex-

es binds more strongly to chiral DNA than does the other enantiomer (28, 29). In luminescence quenching of racemic mixtures of rare earth complexes, resolved ML, complexes stereoselectively quench one of the rare earth species over the other (30, 31). Such chiral recognition promises to be a fundamental and practical tool in spectroscopy and biochemistry. Conclusion Advances in understanding the photophysics and photochemistry of transition metal complexes offer opportunities to use these materials as luminescence sensors and probes. Using a relatively small number of criteria, one can rationalize the "illogical" behavior of the luminescence of metal complexes to provide a framework for the design of new and specialized sensors and probes. We envision several significant areas for future research developments. In the search for specific molecular probes, the design of features that allow site- selective binding while still preserving the optimum luminescence properties is still in its infancy. Adequate pH and CO, sensors of this type still remain to be developed. A deeper understanding of the interactions of the complexes with the substrate or support is still needed and should greatly enhance the design of new probes and sensors. The authors gratefully acknowledge support from the National Science Foundation under grant CHE 88- 17809.

References

I The MB Series offer you the parameters for productive FT-IR : Performance*: sensitivity to 50pA p-p, baseline accuracy of better than 300pA and linearity up to 3A. Constant performance results in analytical reliability! Ease-of-use : fast accurate sample placement, permanently aligned optics and easy software. Minimal training and skill are required for basic operation. ces of software; Basic, Advanced ble resolution up to 1 cm-l and choice of detectors including PAS. Reliability : No down time in 18 months of use for over 90% of owners. Consider the Bomem ME Series. Superior FT-IR at slightly higher price.

Compare with other models: BllMEM oigilab Nicolet PerldnMB100 FIS-7 205 Elmer16PC Yes -

Permanent alignment Exchangeablesample compartment Yes 4 port optiil design Yes ODeratein ally 'orientalionYes Fast detector interchange Yes Smallfootpnnt Yes PC-AT computer Yes Side port Yes Directacquisitionin LabCalc Yes Microscope interface Yes

* On open beam spectrum when measured at4 cm' resolution over 2,1002,000 cm with 20 minutes coaddino using the standard DTGS rlrtprtor

' Models & 5

._ -

_ _ _ ___ _ _ _ _ _ _

--s

I g---

- --- -

MB102 :5,000cm-1to 210 cm-1 MB155 :14,W cm-1to 450 cm' 'Braun MBIOO: 6,000cm-l to 350 cm-l MB160 :15,000cm-l to 3,700 cn SEE US AT THE E A 9 1 MEETING, OCTOBER 28-31, 1991, Booth 3516 CIRCLE 14 ON READER SERVICE CARD

~

--_ _ __ --

Bomem Inc./nanmann ll Braun 450 bnt-Jean-Baptiste Ave Quebec, Quebec

Canada G2E 5%

Tel (418) 877-2944 North America : 1 800 888 FIR

(1)Chemical, Biochemical, and Environmental Sensors; Lieberman, R. A.; Wlodarczyk, M. T., Eds.; The International Society for Optical Engineering: Bellingham, WA, 1989;Vol. 1172. (2) Diamandis, E. P.; Christopoulos, T. K. Anal. Chem. 1990,62,1149 A. (3)Wolfbeis, 0. S. In Molecular Lumines-

cence Spectroscopy Methods and Applications: Part 2; Schulman, S. G., Ed.; John Wiley and Sons: New York, 1988;p. 283. (4)Angel, S. M. Spectroscopy 1989,2(4), 37. (5)J. Chem. Fduc., Oct. 1983,Vol. 60,No. 10 (entire issue). (6) Juris, A.; Balzani, V.; Barigellitti, F.; Campagna, S.; Belser, P.; von Zelewsky, A. Coord. Chem. Reu. 1988,84,85. (7)Demas, J. N.; Crosby, G . A. J. Am. Chem. SOC.1971,93,2841. ( 8 ) Cotton, F.A.; Wilkinson, G. Advanced Inorganic Chemistry; J o h n Wiley a n d Sons: New York, 1980;Chapter 20. (9)Caspar, J. V.; Kober, E. M.; Sullivan, B. P.; Meyer, T. J . J. Am. Chem. SOC. 1982,104,630. (10)Caspar, J. V.; Meyer, T. J. J. Phys. Chem. 1983,87,952. (11) Sacksteder, L.;Zipp, A. P.; Brown, E.; Streich, J.; Demas, J. N.; DeGraff, B. Inorg. Chem. 1990,29,4335.

-

(12) Carstens, D.H.W. Ph.D. Dissertation, University of New Mexico, Albuquerque, 1969. (13) Demas, J. N.; Crosby, G. A. J. Am. Chem. SOC.1970,92, 7262. (14) van Houten, J.; Watts, R. J. J Am. Chem. SOC.1976, 98, 4853. (15) Caspar, J. V.; Meyer, T. J. J. Am. Chem. SOC.1983,105, 5583. (16) Dressick, W. J.; Cline, J., 111; Demas, J. N.; DeGraff, B. A. J. Am. Chem. SOC. 1986, 108, 7567. (17) Kober, E. M.; Marshall, J. L.; Dressick, W. J.; Sullivan, B. P.; Caspar, J. V.; Meyer, T. J. Inorg. Chem. 1985,24, 2755. (18) Dressick, W. J.; Raney, K. W.; Demas, J. N.; DeGraff, B. A. Znorg. Chem. 1984, 23, 875. (19) Reitz, G. A.; Demas, J. N.; DeGraff, B. A.; Stephens, E. M. J. Am. Chem. SOC.

Graff, B. A.; Bacon, J. R. Anal. Chem. 1991, 63, 337. (27) Carraway, E. R.; Demas, J. N. Langmuzr, in press. (28) Kumar, C. V.; Barton, J. K.; Turro, N. J. J. Am. Chem. SOC.1985, 107, 5518. (29) Kirsch-De Mesmaeker, A.; Orellana, G.; Barton, J. K.; Turro, N. J. Photochem. Photobiol. 1990, 52, 461. (30) Metcalf, D. H.; Snyder, S. W.; Demas, J. N.; Richardson, F. S. J. Am. Chem. SOC.1990, 112,469. (31) Metcalf, D. H.; Snyder, S. W.; Demas, J. N.; Richardson, F. S. J. Am. Chem. SOC.1990,112, 5681.

physics, and photochemistry of inorganic metal complexes and related applications to Practical problems. He is involved in the design and development of luminescence instrumentation and methods as well as the application of computers to chemical education. He is a very enthusiastic skier and a third-degree black belt in tae kwon do karate.

1988,110,5051.

(20) Peterson, S. H.; Demas, J. N. J. Am. Chem. SOC.1979,101, 6571. (21) Lakowicz, J. R. Principles of Fluorescence Spectroscopy; Plenum Press: New York, 1983. (22) Harrigan, R. W.; Hager, G. D.; Crosby, G. A. Chem. Phys. Lett. 1973,21, 487. (23) Lippitsch, M. E.; Pusterhofer, J.; Leiner, M.J.P.; Wolfbeis, 0. S. Anal. Chim. Acta 1988, 205, 1. (24) Wolfbeis, 0. S.; Weis, L. J.; Leiner, M.J.P.; Ziegler, W. E. Anal. Chem. 1988, 60, 2028. (25) Bacon, J. R.; Demas, J. N. Anal. Chem. 1987,59, 2780. (26) Carraway, E. R.; Demas, J. N.; De-

B. A. DeGrafreceived his Ph.D. in physical chemistry from The Ohio State University in 1965. He was a postdoctoralfellow at Harvard prior to joining the faculty ofJames Madison University in 1972. His research centers on photochemistry and kinetics of inorganic and organic systems. He is also interested in the development of laser-based experiments and projects for undergraduate education. He enjoys skiing, hiking, and tennis.

J. N. Demas received a Ph.D. in physical chemistry from the University of New Mexico in 1970. He was an NSFpostdoctoral fellow at the University of Southern California Prior to joining the faculty at the University of Virginia in 1971. His research interests include synthesis, photo-

Imagine...

Oxidations in Organic

4

L

T

his volume is the best and most

pounds. It gives a comprehensive, up- 5 to-date, and well-organized review of the subject, with emphasis on preparative aspects and results. An invaluable aide for the practicing chemist, this volume offers a more experimental rather than theoretical scope with emphasis on the prepara- g I tive aspects and synthetic usefulness 4 of individual reactions. Reactions described in the text were chosen on the basis of simplicity,clarity of description in the primary literature, availability of oxidents. and yield. Topics within the volume describe oxidation and dehydrogenation agents and oxidations of various functional groups, with descriptions of the best reagents for a specific reation. The information is expertly organized and supported by examples, correlation tables, and thorough referencing. Milo5 Hudlicky ACS Monograph No. 186 ISBN 0-8412-1780-7 LC 90-34564

ISBN 0-8412-1781-5 LC 90-34564

433 pages (1990) Clothbound $89.95 433 pages (1 990) Paperbound $49.95

-

...performing voltammetry with a computercontrolled instrument that packages superior sensitivity and program-inclusive data manipulation. Imagine this instrument with a totally graphic user interface so straightforward and complete that any researcher, regardless of related discipline or previous experience,can quickly learn its operation and stay current even with infrequent use. Now imagine this instrument optimized for microelectrodes. You've specified our CS-1090 ComputerControlled Potentiostat,an instrument we've crafted for quality, sensitivity and ease of use. For a brochure and demonstration software package call us toll-free. ,

I -800=235=C HEM

cs-1090 Computer Controlled Potentiostat P.O. ~ 0 x 3 9 3 1Lawrence,KS USA 66046 913-#2-2511

5 rq

r 7 3 Cypress Systems Inc.

CIRCLE 24 ON READER SERVICE CARD

ANALYTICAL CHEMISTRY, VOL. 63, NO. 17, SEPTEMBER 1,1991

837 A