An EPR Study of Two-Electron Sensitization by Fragmentable Electron

10126

J. Phys. Chem. B 2005, 109, 10126-10136

An EPR Study of Two-Electron Sensitization by Fragmentable Electron Donors R. S. Eachus, A. A. Muenter,* T. D. Pawlik, and J. R. Lenhard Research and DeVelopment Laboratories, Eastman Kodak Company, 1999 Lake AVenue, Rochester, New York 14650-1735 ReceiVed: January 26, 2005; In Final Form: March 28, 2005

Fragmentable electron donors (FEDs) are molecules designed to undergo bond fragmentation after capturing the hole created by photoexposure of silver halide. By design, the radical remaining after fragmentation is a potent reductant expected to be capable of injecting an electron into the conduction band of the silver halide. Thus, the addition of a FED to the AgX surface should allow the creation of two electrons for each photon absorbed by the substrate. Photographic studies have confirmed that the addition of FEDs can increase the photosensitivity of AgX materials.1 In this work, EPR spectroscopy was employed to study the processes of hole capture, donor fragmentation, and secondary electron injection by FEDs in AgBr dispersions. To do so, we used AgBr microcrystals doped with diamagnetic transition metal complexes that act as deep electron traps. For samples exposed to actinic light at 15 K, secondary electron injection was detected as an increase in the EPR signal from electrons trapped at the dopant upon annealing the samples above 50 K. Organic radical intermediates and self-trapped hole centers were the other paramagnetic species monitored in this study. The results presented here confirm that the FED sensitization mechanisms originally proposed by Gould et al.1 take place at silver halide surfaces and result in additional electrons in the silver halide conduction band.

Introduction A method to substantially increase the imaging efficiency of state-of-the-art photographic materials has been reported by Gould et al.1 and described in several patents.2-5 This novel approach is summarized in Scheme 1. SCHEME 1

The photographic system, described as AgBr|Dye in eq 1, consists of silver halide microcrystals (or grains) dispersed in gelatin. These grains are spectrally sensitized with an adsorbed dye. The novel aspect of this scheme entails the co-addition of an electron-donor molecule, XY. Light absorption by the dye initiates the transfer of an electron to the conduction band of the silver halide substrate,6 a condition represented here as (AgBr + e-). This is the first step in latent image formation.6 The resultant oxidized dye radical, Dye•+, reacts with XY to give a radical cation, XY•+, which then fragments to a radical X• and a stable cation Y+. XY is described as a fragmentable electron donor, or FED. Its structure is designed so that X• is a powerful reductant and has the potential to transfer a second electron into the silver halide conduction band. Thus, two electrons are injected per absorbed photon, effectively doubling the imaging efficiency of the system. This process is referred to as twoelectron sensitization. * Address correspondence to this author. Address: 9 Park Place, Rochester, NY 14625. Phone: (585) 381-2450. E-mail: amuenter@ rochester.rr.com.

There is a potential additional advantage to the co-addition of XY and the dye. The reverse of eq 1, i.e., recombination of the injected electron with an oxidized dye radical at the grain’s surface, can limit the imaging efficiency of the system. In this situation, the rapid reaction of Dye•+ with XY (eq 2) should suppress this loss process and, thereby, increase the system’s imaging efficiency, even if X• is unable to subsequently inject a second electron. It is also possible for the FED compound to function in the absence of a sensitizing dye. In this case, the photon is absorbed directly as a band-gap excitation of the silver halide, creating an electron and a hole. The XY moiety captures the photohole, and the fragmentation and electron injection steps occur as described above. Photographic measurements on both undyed and dyed silver halide coatings containing a range of FED molecules have confirmed significant sensitivity increases from the addition of the FED compounds. Solution transient absorption studies of related model systems have indirectly supported the proposed two-electron mechanism for this sensitivity increase.1-5 Nevertheless, definite verification that the twoelectron mechanism can function for compounds adsorbed at a silver halide surface and can produce conduction band electrons was lacking. The photographic sensitivity increases obtained with FED molecules are not in themselves definite evidence, since it is possible to hypothesize a dark reaction of the compounds that produces small silver clusters that act as sensitivity centers. This type of sensitivity increase mechanism is termed reduction sensitization.7 In this report, we present the first direct evidence for the operation of the two-electron mechanism at a silver halide surface, based upon magnetic resonance measurements on silver bromide dispersions. Experimental Strategy Because the intermediates Dye•+, XY•+, and X• are paramagnetic, photo-EPR (electron paramagnetic resonance) spec-

10.1021/jp0504672 CCC: $30.25 © 2005 American Chemical Society Published on Web 05/04/2005

An EPR Study of Two-Electron Sensitization troscopy is an appropriate technique to test Scheme 1. Transient optical absorption data from ref 1 suggest that these radicals are likely to be very short-lived in photographic dispersions at room temperature. To stabilize them sufficiently to facilitate continuous wave EPR studies, irradiations and measurements have been performed at lower temperatures. To minimize recombination and to test for secondary electron injection, the silver bromide grains were doped internally with one of several transition-metal complexes known to function as deep electron traps. The best dopants for this purpose proved to be the diamagnetic anions [Cl5Rh(H2O)]2- and [Cl5Os(NO)]2-.8-10 The corresponding [Cl5Rh(H2O)]3- and [Cl5Os(NO)]3- trapped electron centers are long-lived in AgBr, and their EPR spectra are known. The radical intermediates of interest are formed at, or near to, surfaces of the silver halide grains. Thus, their concentrations should increase with the surface-to-volume ratio of the substrate. For this reason, our experiments have been confined to cubic AgBr grains with small edge lengths, typically 0.1 ( 0.01 µm. This has the additional advantage of providing optimum ensemble averaging for the powder-like EPR spectra obtained from these dispersions, leading to narrow lines and enhanced detection sensitivity. The samples examined in this study were prepared as 1 mm thick coatings by evaporation of freshly prepared AgBr dispersions. The resultant semiopaque materials contained large amounts of gelatin and were thereby contaminated with several paramagnetic ions, particularly Cu2+, Fe2+, Fe3+, and Mn2+. Fortunately, these impurities are not photoactive in gelatin and so their spectra could be subtracted to reveal signals from the paramagnetic species generated by actinic radiation. All of the spectra described here were obtained in this manner. For a FED molecule to function as a two-electron sensitizer in the silver halide system, three requirements are necessary: the oxidation potential of the parent molecule must be low enough to allow it to capture the photohole created in the silver halide or in the dye; the fragmentation reaction must occur within the time scale of silver halide latent image formation; and the radical created by the fragmentation must be sufficiently reducing to inject an electron into the silver halide conduction band. The FED molecules selected for this study were amino acids that, upon oxidation by a photohole or by Dye•+, are expected to give intermediates that can rapidly fragment by decarboxylation to R-amino radicals with strongly negative oxidation potentials.1

J. Phys. Chem. B, Vol. 109, No. 20, 2005 10127 condition required for the radical to easily inject an electron into the AgBr conduction band. A further requirement for the FED compounds to be useful sensitizers is that they should be adsorbed to the silver halide surface. While this can be accomplished by adding high concentrations of the simple amino acid structures to silver halide dispersions, it is preferable to use compounds where a covalent linkage to an adsorbing moiety is provided.2,5 Structures for the two FED examples employed in the majority of this work are shown below. In these compounds, R2 is a linking bridge to a phenyl mercaptotetrazole, a moiety that is known to adsorb strongly to silver halide by complexation of the mercapto group with silver ion.11

In PMT-3, the R1 formyl substituent leads to a relatively high oxidation potential for the FED moiety, approximately 1.18 V, and a very rapid room temperature solution fragmentation rate, probably near 1011 s-1. In PMT-41, with H as R1 in the FED moiety, the estimated oxidation potential is lower, 0.91 V, and the room temperature fragmentation rate is slower, roughly 108 s-1. These fast fragmentation reactions of the radical cations XY•+ make an exact experimental measurement of the formal oxidation potentials of the FED moieties extremely difficult to obtain, and the estimates of the potentials have been obtained by using model compounds, as described in the Experimental Section. Since cyanine dyes with oxidation potentials of 1.5 V have been shown to have their ground states approximately aligned with the AgBr valence band,12 both of these FED compounds should be able to trap photoholes from the AgBr substrate. The substituted phenyl mercaptotetrazole a-PMT was used

SCHEME 2

In the amino acid structure shown in Scheme 2 above, the oxidation potential of the parent compound can be varied by changing the substituents R1 and R2. The fragmentation rate after oxidation is also strongly correlated with this oxidation potential, with compounds having lower oxidation potentials exhibiting slower fragmentation rates. The substituent R3 influences the reducing power of the radical remaining after fragmentation. Radicals where R3 is methyl have been found to have oxidation potentials more negative than -0.9 V, the

as a reference compound for these experiments. This material is expected to complex with the silver halide surface in a manner similar to that of PMT-3 or PMT-41, but given its much higher oxidation potential (approximately 2.1 V), it is not expected to trap photoholes. Thus, it should give information on whether the surface complexation itself affects the EPR signals obtained. In addition, the oxidized form of the compound does not fragment. The sensitizing dye employed in the experiments was the well-known blue-absorbing, anionic cyanine (BTHC).

10128 J. Phys. Chem. B, Vol. 109, No. 20, 2005

Figure 1. A 9.3-GHz EPR spectrum obtained at 15 K, following the exposure of an evaporated AgBr dispersion to actinic light. The AgBr grains were doped with 100 molar ppm of [Cl5Rh(H2O)]2-. This dopant proved to be contaminated with traces of [Cl4Rh(H2O)2]1- and (RhCl6)3-. The solid line is the usual first-derivative presentation; the dotted line is its integral. The region shaded gray is attributed to [(Br4)3-‚V] trapped hole centers; see text for details.

This dye readily adsorbs to AgBr grains forming a J-aggregate with peak absorption at about 470 nm.13 Its oxidation potential is 1.39 V;14 therefore, it is expected that holes trapped at this dye can transfer to any of the FEDs studied here. As an alternate method of associating a FED with the silver halide surface, the FED moiety can be covalently linked to the sensitizing dye. For some experiments, an example containing the BTHC chromophore was employed as shown below.

The R1 methyl group in the FED moiety of DFED-11 results in an estimated oxidation potential of 0.77 V for this material and a room temperature fragmentation rate of about 107 s-1. The control comparison for this compound was its ethyl ester, DFED-11E, where esterification blocks decarboxylation and, therefore, two-electron sensitization behavior. Results and Discussion Undyed Rh3+-Doped AgBr. (i) Without a-PMT or FED. Figure 1 shows a 9.3-GHz EPR spectrum obtained from an evaporated dispersion of AgBr in which the 0.1 µm cubic grains were doped nominally with 100 molar parts per million of [Cl5Rh(H2O)]2-. The sample was undyed and free of a-PMT or FED. This spectrum was measured with the sample at 15 K, following its exposure to actinic light at that temperature. Identical results were obtained for exposures between 365 and 440 nm. The spectrum is presented in the usual first-derivative mode as the solid line, while the dotted curve is its integral. This and subsequent spectra were measured at 50 µw microwave power, a condition that avoided signal saturation. Thus, signal

Eachus et al. intensity is directly related to radical concentration. There are features from at least four photogenerated species convoluted in this spectrum. They can be assigned by inspection from a comparison to literature data. The region of the signal centered near 320 mT (shaded gray in the first derivative curve in the figure) is from the intrinsic hole center [(Br4)3-‚V].9 This defect comprises a hole shared by four equivalent bromide ions in a planar array around a silver ion vacancy (V). It is located at, or near, the grain’s surface. The weak, narrower signal near 330 mT results from a small population of silver atom clusters, (Ag0)n, oxidized gelatin radicals, gel+, or a mixture of these species.15 The strongest feature in the derivative spectrum is from the trapped electron center [Cl5Rh(H2O)]3-. This complex has orthorhombic or lower symmetry, but gx and gy differ by less than 0.01 so that they cannot be resolved in this X-band spectrum. They are resolved at Q-band (35 GHz), however. At X-band, they are convoluted into the perpendicular-like feature at 264 mT (g ) 2.52). The expected feature corresponding to gz (g ≈ 2.0) is broad, weak, and convoluted with the intrinsic hole signal around 330 mT. The fourth species gives the weak, low-field shoulder near 257 mT (g ) 2.58). It is assigned as [Cl4Rh(H2O)2]2- formed by photoelectron trapping. The complex [Cl4Rh(H2O)2]1- has proved to be an unavoidable impurity in aqueous solutions of [Cl5Rh(H2O)]2-. In some samples, electron trapping by the additional impurity (RhCl6)3- could be detected as a perpendicular-like feature at 273 mT (g ) 2.43), especially in spectra measured at Q-band. The first derivative presentation in Figure 1 makes it difficult to estimate the relative concentrations of the four photoproducts. Inspection suggests that the yield of trapped electron centers is substantially greater than that of [(Br4)3-‚V], but the integrated signal shows that they are actually produced in about equal concentrations. To obtain metrics for the concentrations of Rh2+ trapped electron centers, Ae, and [(Br4)3-‚V] intrinsic hole centers, Ah, the first-derivative EPR spectrum in Figure 1 was doubly integrated. (Note that Rh2+ is used here to denote a combination of [Cl5Rh(H2O)]3-, [Cl4Rh(H2O)2]2-, and (RhCl6)4-.) The Rh2+ signal in the region of “g⊥” is substantially shifted downfield from g ) 2. Thus, the area between 250 and 280 mT gives a good measure of Ae. On the other hand, the high-field region of the spectrum from [(Br4)3-‚V] strongly overlaps those derived from several organic radicals observed in this study. To obtain a consistent measure of Ah, we have only used the area under the signal between 300 and 325 mT. In these experiments, Ae and Ah reached saturation values after about 1400 s of exposure to 365-440 nm light at 15 K. This slow filling rate is a result of the opacity of the gelatin dispersion combined with recombination losses. In subsequent experiments, exposure times were limited to 660 s to retain a sufficient population of vacant Rh3+ centers as detectors for secondary electron injection from X• radicals. Annealing experiments were carried out by heating the irradiated dispersion to a given temperature in the dark, holding for 10 min, and quenching back to 15 K for measurement of the modified EPR signal. The plots of normalized EPR signal intensity vs annealing temperature shown in Figure 2 illustrate the changes observed. For convenience in separating the two signals in this and subsequent plots, the Ae signal is normalized to 1.0 at 15 K, while the Ah signal is normalized to 0.5. Both Ae and Ah decrease with increasing annealing temperature. It is known that [(Br4)3-‚V] centers are far less stable than Rh2+ in AgBr,8,10 and that they decay by thermal excitation of the trapped hole to the AgBr valence band edge. Some of

An EPR Study of Two-Electron Sensitization

Figure 2. Plots of signal intensity vs annealing temperature for a [Cl5Rh(H2O)]2--doped AgBr dispersion. The sample was initially exposed to 365-440 nm light at 15 K. See text for annealing procedure. Ae represents the concentration of electrons trapped internally at Rh3+ centers. Ah represents the concentration of [(Br4)3-‚V] trapped hole centers at the AgBr grain surface.

the resultant mobile holes react with the external peptizing medium to form gel+ radicals,15 and some recombine with Rh2+, resulting in the observed decrease in the trapped electron signal. (ii) With a-PMT. The addition of 2 mmol/mol of AgBr of the nonfragmenting compound a-PMT affected the photochemistry of the dispersion. For band-gap exposures at 15 K, the formation of Rh2+ centers was basically unchanged by the addition of this compound, but the EPR signal from [(Br4)3-‚V] hole centers was shifted to slightly higher field (lower g factor). Annealing to 90 K yielded a hole center signal that was further shifted toward high field, suggesting the coexistence of two overlapping signals, with the higher field one being associated with the presence of a-PMT and having somewhat greater thermal stability. It is possible that this modification in the [(Br4)3-‚V] spectrum reflects a change in the surface structure of the AgBr grains. a-PMT is known to bind strongly to Ag+ ions, and it suppresses cation interstitial formation at AgBr surfaces.11 A perturbation of the surface resulting from a-PMT adsorption could cause a structural distortion of [(Br4)3-‚V], with corresponding changes in the hole distribution, its g factors, and its thermal stability. The thermal properties of the Rh2+ centers were also mildly affected by a-PMT addition, as illustrated by the annealing data in Figure 3. Recombination of mobile holes with Rh2+ in the temperature range 150-275 K was suppressed, suggesting that this recombination is a less important decay pathway for the a-PMT modified hole center. (iii) With PMT-41. The X-band EPR spectrum in Figure 4 was obtained at 15 K from an AgBr dispersion in which the [Cl5Rh(H2O)]2--doped cubic grains had the FED compound PMT-41 added at a concentration of 2.0 mmol/mol of AgBr. The dispersion was exposed to band-gap radiation for 660 s at 15 K. Signals from Rh2+-trapped electron centers are evident in the region between 250 and 280 mT, and an intense single line at about 331.5 mT (∆Bp-p ) 3.3 mT) is observed. This line results from a new radical. Formation of the intrinsic hole center [(Br4)3-‚V] is completely suppressed at this FED concentration, although it was still detected in dispersions treated with e0.3 mmol/mol of AgBr of PMT-41. This suggests that the new signal in Figure 4 results from a trapped hole center such as an oxidized PMT-41•+ radical or its decomposition product. The spectrum was doubly integrated over the magnetic field ranges

J. Phys. Chem. B, Vol. 109, No. 20, 2005 10129

Figure 3. Plots of signal intensity vs annealing temperature for [Cl5Rh(H2O)]2--doped AgBr dispersions with and without treatment with 2 mmol/mol of AgBr of the nonfragmenting donor a-PMT. Ae represents the concentration of electrons trapped internally at Rh3+ centers. Ah represents the concentration of [(Br4)3-‚V] hole centers trapped at grain surfaces.

Figure 4. EPR (9.3 GHz) spectra obtained at 15 K following the exposure of an evaporated AgBr dispersion to 365-440 nm light. The AgBr grains were doped with 100 molar ppm of [Cl5Rh(H2O)]2-. PMT41 was added at a concentration of 2 mmol/mol of AgBr.

indicated in the figure to obtain rough measures of concentration for PMT-41•+ radicals, AR, and trapped electron centers, Ae. The concentration of PMT-41•+ radicals initially formed after illumination increased as the surface concentration of PMT-41 was increased from 0.3 mmol/mol of AgBr to 1.0 mmol/mol of AgBr. Once again, annealing the irradiated dispersion for 10 min in the dark caused significant changes in the EPR spectrum obtained after quenching back to 15 K. Figure 5 compares annealing data obtained from dispersions with three different concentrations of the FED added. Values for Ae and Ah from the sample containing no FED are included for reference, as well as Ah from the sample treated with 0.3 mmol/mol of AgBr of PMT-41. Values for AR are normalized to 1.0 at 15 K. There was an initial rise in the PMT-41•+ radical concentration upon annealing to 90 K. This increase was largest for the dispersion with the lowest level of this FED. In this sample, AR reached a maximum value at about 120 K, by which temperature any intrinsic hole centers had decayed. This correlation indicates that one of the decay pathways for the [(Br)43-‚V] centers is transfer of the hole to the PMT-41, either directly at the surface or indirectly via the AgBr valence band. The absence of detectable concentrations of [(Br)43-‚V] centers in the dispersions treated with higher levels of the FED accounts for the smaller increases in PMT-41•+ radical concentrations obtained by annealing.


Figure 5. Plots of signal intensity vs annealing temperature for four [Cl5Rh(H2O)]2--doped AgBr dispersions: (i) no surface additions; (ii) PMT-41 at a concentration of 0.3 mmol/mol of AgBr; (iii) PMT-41 at a concentration of 1.0 mmol/mol of AgBr; and (iv) PMT-41 at a concentration of 2.0 mmol/mol of AgBr. Ae represents the concentration of Rh2+ centers, Ah represents the concentration of [(Br4)3-‚V] hole centers, and AR represents the concentration of PMT-41•+ radicals. The addition of PMT-41 at concentrations above 0.3 mmol/mol of AgBr suppressed the formation of [(Br4)3-‚V] hole centers to concentrations below the EPR detection limit.

Figure 6. ENDOR spectra of a [RhCl5(H2O)]2-doped AgBr dispersion with 1 mmol/mol of AgBr of PMT-41 added: trace a before exposure and trace b after light exposure (365-440 nm) at 100 K for 30 min. The ENDOR measurement was performed at 14 K with the static magnetic field set to 336.74 mT corresponding to a g value of 2.004.

Without the addition of PMT-41, the concentration of Rh2+ electron traps decreased on annealing until, at 300 K, about 40% of the trapped electron centers survived. The addition of PMT41 suppressed recombination at the Rh2+ centers and, at a level of 1 mmol/mol of AgBr, actually produced an increase in Rh2+ level after annealing above 160 K. At 300 K, the Rh2+ signals increased to about 150% of their original intensities. This increase in the Rh2+ signal is accompanied by decay of the PMT-41•+ radical between 160 and 300 K. This result indicates secondary electron injection into the AgBr substrate by the PMT-41•+ radical, either directly or via an intermediate decomposition product. A further increase of the PMT-41 level to 2 mmol/mol did not lead to an additional increase in Rh2+ signal. The strong EPR signal obtained from PMT-41•+ radicals made ENDOR measurements feasible. Figure 6 shows the powder-ENDOR spectrum obtained from a [RhCl5(H2O)]2doped AgBr dispersion with 1 mmol/mol of AgBr of PMT-41 added to the grain surface. Trace a was recorded before light exposure. Trace b is the ENDOR spectrum after band-gap exposure for 30 min at 100

Eachus et al.

Figure 7. Plots of signal intensity vs annealing temperature for three [Cl5Rh(H2O)]2--doped AgBr dispersions: (i) no surface additions, (ii) PMT-3 at a concentration of 2.0 mmol/mol of AgBr, and (iii) PMT-41 at a concentration of 2.0 mmol/mol of AgBr. Ae (closed symbols) represents the concentration of Rh2+ centers, Ah (open symbols, dotted lines) represents the concentration of [(Br4)3-‚V] hole centers, and AR (open symbols, solid lines) represents the concentration of PMT-3•+ or PMT-41•+ radicals.

K. The magnetic field was set to the center of the PMT-41•+ EPR line. The features in trace b marked with arrows correspond to 1H hyperfine interactions. They are spaced symmetrically about the 1H Larmor line at 14.34 MHz. Their separation corresponds to a hyperfine interaction of 6 MHz. This value is small enough that the resultant splittings in the EPR spectrum are within the experimental line width and, therefore, unresolved. Ab initio calculations described below suggest that this intermediate has the N-centered structure:

This is the unfragmented form of the PMT-41•+ radical. The absence of resolved 14N hyperfine features in the ENDOR spectrum is not unexpected. Nuclear quadrupole interactions frequently broaden such features beyond detection in powder spectra.16 (iV) With PMT-3. Similar results to those reported for PMT41 were obtained for PMT-3. The addition of this FED compound at a concentration of 2 mmol/mol to a Rh3+-doped AgBr dispersion decreased the formation of intrinsic hole centers during band-gap irradiation at 15 K. This suppression, the result of the competitive formation of oxidized FED radicals, was similar to that seen for the 0.3 mmol/mol level of PMT-41. The appropriate concentration of Rh2+-trapped electron centers was also produced. The PMT-3 radicals were detected by 9.3-GHz EPR spectroscopy as a single line centered at 331.5 mT with ∆Bp-p ) 2.6 mT. The effects of PMT-3 were also evident in annealing experiments. Figure 7 compares data from dispersions with no surface treatment and with 2.0 mmol/mol of AgBr of PMT-3 or PMT-41 added. Although their effects are similar, PMT-3 is a less efficient hole trap than PMT-41. Both compounds should be deep hole traps at 15 K; therefore, this apparent difference in trapping cross-section may be related to differences in geometry of the two compounds as adsorbed at the AgBr surface. In addition, PMT-3•+ is observed to be less stable than PMT-41•+: the disappearance of PMT-3•+ begins at roughly 90 K vs 140 K


Figure 8. A 9.3 GHz EPR spectrum obtained at 15 K following the exposure of an evaporated Rh3+-doped AgBr dispersion to 365-440 nm light at 15 K (dotted curve) and after annealing in the dark at 160 K for 10 min (solid curve). Only the central region of the spectrum is shown here. The grains were treated with PMT-3 at a concentration of 1.0 mmol/mol AgBr. The arrows mark potential 1H and/or 14N hyperfine features.

for PMT-41•+. This stability difference is consistent with the higher rate of fragmentation observed in a room temperature solution for the oxidized PMT-3 FED moiety. As in the case of PMT-41•+, disappearance of PMT-3•+ was accompanied by a significant decrease in the loss of Rh2+ with increasing temperature. These results support the contention that, like PMT41•+, the PMT-3•+ radical is a source of secondary electrons in AgBr dispersions. We propose that PMT-3•+ is an N-centered radical with a structure analogous to that shown above for PMT41•+. In the course of PMT-3•+ decomposition, a weak signal from a new radical was observed (see Figure 8), its concentration reaching a maximum value after annealing at 160 K for about 10 min. This radical was not detected following annealing at temperatures of 200 K and higher. The signal is weak, but it is possible to resolve splittings of about 2 mT in the wings (marked with arrows in the figure). These might be 1H and/or 14N hyperfine splittings. PowderENDOR measurements have revealed no further structural information about this species, but estimates of 1H hyperfine interactions from ab initio calculations (see below) are consistent with this signal originating from the decarboxylated radical shown below:

This observation provides support for the identification of this radical as the intermediate that injects the secondary electron into the AgBr conduction band, as indicated in Scheme 2. The corresponding radical remaining after decarboxylation of PMT41•+ was not observable, presumably because, at the higher temperature required to significantly fragment PMT-41•+, the decarboxylated radical was not stable enough to build up a


Figure 9. EPR (9.3 GHz) spectra obtained at 15 K, following the exposure of evaporated AgBr dispersions to 440-480 nm light. The AgBr grains were doped with 100 molar ppm of [Cl5Rh(H2O)]2-. Light curve: the spectrum shown in Figure 1 from a dispersion containing no dye or FED. Dark curve: the spectrum from grains sensitized with a blue cyanine dye.

detectable steady-state concentration. Decarboxylation of both PMT-3•+ and PMT-41•+ yields R-amino carbon radicals with methyl substitution; therefore these two decarboxylated radicals are expected to have similar stability. The fact that the decarboxylated radical from PMT-3•+ exhibits moderate stability at 160 K suggests that the process of electron injection from these radicals into the AgBr conduction band is thermally activated. The EPR data described above for the undyed, Rh3+-doped AgX dispersions support the conclusion that the proposed twoelectron sensitization mechanism for FED compounds takes place at the AgX surface, providing additional electrons in the AgX conduction band. The experimental approach of exposure at low temperature and subsequent thermal annealing allows the concentration of all the relevant radicals to be monitored by EPR and helps to verify that the steps in the two-electron sensitization process can take place at a silver halide surface. Because the Rh2+-trapped electron centers have reasonable stability at room temperature, we also examined the concentration of these centers formed by illuminating the doped dispersions at room temperature, then cooling to 15 K to perform the EPR measurements. For the dispersions containing PMT-41 or PMT-3, exposures at 440 nm for 2160 s produced concentrations of Rh2+-trapped electron centers that were 2.5 to 3 times larger than the concentration seen in the dispersion with no FED added. These room temperature observations are further support for the functioning of the two-electron sensitization mechanism at the AgX surface. Dye-Sensitized, Rh3+-Doped AgBr. (i) With Dye Alone. For any practical application of this technology, the AgBr grains will be optically sensitized with an appropriate dye. To assess the effects of adding the FED in combination with a dye, the blue-absorbing, anionic cyanine dye BTHC was added to the system. This molecule readily adsorbs to AgBr grains, forming a J- aggregate with a peak absorption at about 470 nm.13 The effects of optical sensitization on the photochemistry of Rh3+-doped AgBr grains are illustrated by the EPR data in Figure 9. Wavelengths between 440 and 480 nm gave identical results for these light exposures at 15 K. There are strong signals from Rh2+-trapped electron centers and a new feature at 331.5 mT (g ) 2.0047; ∆Bp-p ) 1.8 mT)


Figure 10. A plot of signal intensity vs annealing temperature for two Rh3+-doped AgBr dispersions. One has sensitizing dye added to the dispersion. AD represents the concentration of Dye•+ radicals produced by 660 s exposure to 440-480 nm light at 15 K. Ae represents the concentration of electrons trapped internally at Rh3+ centers.

assigned to oxidized dye radicals Dye•+. The presence of the J-aggregated cyanine does not completely eliminate the formation of intrinsic [(Br4)3-‚V] centers because 440-480 nm light excites both a HOMO-to-LUMO transition in the dye and weak band-to-band excitation in AgBr. It does, however, reduce the yields of both Rh2+ and [(Br4)3-‚V] produced at 15 Ksprobably the result of recombination at the dye. Once again, annealing the irradiated dispersion for 10 min in the dark caused changes in the EPR spectrum obtained after quenching back to 15 K, as illustrated by the data in Figure 10. Here, AD represents the concentration of dye radicals obtained by double integration of the region in Figure 9 between 325 and 335 mT. This value for AD is normalized to 0.5 at 15 K. In undyed grains, Ae decreased steadily with increasing annealing temperature. The effect of the adsorbed dye was to postpone the decay of Rh2+ centers to higher annealing temperatures, increasing by about 75% the relative amount of this radical remaining at 200 K. We attribute this to the suppression of recombination at the dopant. In fact, there was a 25% increase in AD following annealing at 90 K, the region where [(Br4)3-‚V] centers were bleached, consistent with the reaction of mobile holes with additional dye at the grain’s surface. The decay of Dye•+ radicals was evident above 90100 K, but hole injection back into AgBr did not appear to be a primary loss process. If it was, the increased stability of Rh2+ observed experimentally could not be explained. (ii) With a-PMT. Data plotted in Figure 11 show the effects of a-PMT co-addition with the dye. The concentration of a-PMT was 1.0 mmol/mol of AgBr. This compound slightly increased the stability of the Rh2+trapped electron centers, but accelerated the decay of Dye•+ radicals. This effect is especially evident above 50 K. Figure 12 shows EPR spectra from a dispersion containing dye only and a dispersion with both dye and a-PMT. The spectra were recorded following exposure at 440-480 nm and annealing at 120 K. The intensity of the Rh2+ signal is the same in both spectra, but the intensity of Dye•+ is reduced by 50% in the dispersion containing a-PMT, while a new, broad EPR band appears to low field of the dye signal. This band, circled in Figure 12, is similar to the shifted [(Br4)3-‚V] EPR signal that was observed in the undyed dispersions with a-PMT. We suggest that this signal is again an a-PMT-modified hole center. It is formed when a-PMT-modified surface sites compete

Eachus et al.

Figure 11. A plot of signal intensity vs annealing temperature for three Rh3+-doped AgBr dispersions. One has no surface additions, the second has dye only, and the third was treated with both dye and the reference compound, a-PMT. AD represents the concentration of Dye•+ radicals, while Ae represents the concentration of Rh2+ centers.

Figure 12. EPR (9.3 GHz) spectra obtained at 15 K following the exposure of evaporated AgBr dispersions to 440-480 nm light and annealing in the dark at 120 K for 10 min. The AgBr grains were doped nominally with 100 molar ppm of [Cl5Rh(H2O)]2-. Light curve: spectrum obtained from a dispersion containing dye only. Dark curve: these grains were sensitized with the dye and a-PMT at a concentration of 1.0 mmol/mol of AgBr. The signal circled in the insert is assigned to the [(Br4)3-‚V] center modified by the presence of a-PMT.

with the Dye centers for holes that are thermally released from the unperturbed [(Br4)3-‚V] centers during the annealing step. (iii) With PMT-3. Co-addition of PMT-3 and the dye leads to the formation of both Dye•+ and PMT-3•+ radicals after light exposure. Figure 13 shows EPR spectra in the field region corresponding to g ) 2 after 440-nm exposure at 15 K and subsequent annealing to 120 K for dispersions with dye alone, PMT-3 alone, and dye with PMT-3. It illustrates that the EPR signals of Dye•+ and PMT-3•+ are distinguishable. The EPR signal of Dye•+ is narrower than the EPR signal of PMT-3•+, and the presence of both species is clear in the exposed, annealed dispersion containing the two compounds. The result of a deconvolution of the two signals to yield the annealing temperature dependence of AD and AR for the sample containing both dye and PMT-3 is given in Figure 14. The experiment shows that as the EPR signal from Dye•+ decays at about 100 K, the intensity of the PMT-3•+ radical signal increases by about a factor of 2. This is experimental evidence for step ii in the proposed FED mechanism according to Scheme 1, i.e., the transfer of dye holes to the FED molecule.


Figure 13. Comparison of the EPR spectra of three dispersions after 440 nm of exposure at 15 K and subsequent annealing to 120 K: (i) dye only, (ii) PMT-3 only (1.0 mmol/mol), and (iii) dye + PMT-3 (1.0 mmol/mol).


Figure 15. Plots of signal intensity versus annealing temperature for four Rh3+-doped AgBr dispersions: (i) no surface additions; (ii) dye only; (iii) dye and 0.3 mmol/mol of AgBr of PMT-3; and (iv) dye and 1.0 mmol/mol of AgBr of PMT-3. AD + AR represents the cumulative concentration of Dye•+ and PMT-3•+ radicals (open symbols) and Ae represents the concentration of Rh2+-trapped electron centers (closed symbols).

effect, which increased with the concentration of PMT-3 added to the dispersion, is attributed to secondary electron injection into AgBr, in a manner similar to that observed for PMT-3 on the undyed AgBr grains (Figure 7). In these dyed grains, it was again possible to observe the radical attributed to the decarboxylated PMT-3•+. Both these observations indicate that, whether the photohole is initially created in the AgBr valence band or in a sensitizing dye, the two-electron sensitization process is capable of injecting secondary electrons into the AgBr conduction band. [Cl5Os(NO)]2--Doped AgBr Treated with DFED-11 and Its Ester. An interesting variation of the two-electron sensitization strategy discussed in ref 1 involves the coordination of the FED moiety directly to the dye. The resultant molecule is then adsorbed to the silver halide. The photolysis mechanism envisaged for such a system is represented in Scheme 3. SCHEME 3 Figure 14. Signal intensity vs annealing temperature for a Rh3+-doped AgBr dispersion with both dye and PMT-3. Signals are not normalized in order to more clearly show the result of deconvolution.

The overall influence of the co-addition of Dye and PMT-3 on the stability of Rh2+ centers is shown in Figure 15. It includes annealing data from samples containing PMT-3 at two concentrations, 0.3 and 1.0 mmol/mol of AgBr. In these plots, the signals for Dye•+ and PMT-3•+ have not been deconvoluted and are shown as the combined intensities of the two species, AD + AR. Up to an annealing temperature of about 50 K, the effect of PMT-3 addition on the concentration of trapped electron centers was similar to that of dye alone. This result is explained if some PMT-3 is oxidized directly by intrinsic holes at the AgBr surface, thereby suppressing recombination at Rh2+ centers. PMT-3 did not change the AD + AR radical concentration below 100 K. However, addition of PMT-3 did increase the rate of decay of these radicals above 100 K, as would be expected if dye holes are transferred to PMT-3, which fragments in this temperature range. The most significant influence of PMT-3 addition was observed between 100 and 200 K, where it increased the number of trapped electron centers produced by annealing, compared to samples with dye alone or in combination with a-PMT. This

Our initial work exploring the ability of EPR measurements to probe two-electron sensitization at silver halide surfaces employed this type of dye-linked FED moiety. After the success of this experiment, described below, we moved to the previously discussed, more detailed studies of the PMT-linked FED moieties. Such FED compounds are more tractable synthetically than the dye-linked compounds and thus more closely related to compounds with practical utility. The compound used in this initial study was DFED-11, represented as D-XY in eq 6. To detect conduction band electrons injected from the optically excited sensitizer, D*XY, and from the fragmented donor, D-X•, the AgBr grains were doped internally with [Cl5Os(NO)]2-. This dopant has received attention, both experimentally and theoretically.9 It is incorporated substitutionally into AgBr, and theory predicts that it replaces an [Ag4Br6]2- lattice unit. The low-spin 5d6 ion Os2+ substitutes for Ag+. At room temperature, three negatively charged silver ion vacancies are strongly bound at positions adjacent to the nitrosyl ligand. This ligand, which carries an effective charge of +1, straddles a lattice anion site making it


Eachus et al.

Figure 16. EPR (9.3 GHz) spectra obtained at 15 K following the exposure of evaporated AgBr dispersions to 440-480 nm light. The AgBr grains were doped with 100 molar ppm of [Cl5Os(NO)]2-. (a) Dispersion sensitized with the ester DFED-11E. (b) Dispersion sensitized with DFED-11. Dark curves show spectra obtained immediately following irradiation at 15 K. Light curves show spectra obtained after postexposure annealing to 200 K and quenching back to 15 K. See text for a discussion of assignments.

a potent trap for photoelectrons. One-electron reduction produces the species [Cl5Os(NO)]3-‚3V in which the trapped carrier is essentially localized on the nitrosyl ligand. Once again, V represents a silver ion vacancy. The effective charge of the reduced dopant-vacancy complex is now -1. If photolysis is performed at very low temperatures, the three vacancies remain bound to the dopant, but annealing causes diffusion away of the extra vacancy to a more distant lattice site. This relaxation process stabilizes the trapped electron center as the neutral species [Cl5Os(NO)]3-‚2V. Both electron trapping and ionic relaxation have been predicted theoretically and monitored by EPR spectroscopy.9 Clear differences in g factors are seen for the two dopant-vacancy species. This is the result of -OsNdO bond bending, which is more pronounced for the complex with two bound vacancies. Figure 16 illustrates the integrated spectra obtained after light exposure at 15K for 0.11 µm cubic AgBr grains doped with 100 molar ppm of [Cl5Os(NO)]2- and sensitized with DFED11 or its reference nonfragmenting ester, DFED-11E. As shown in Figure 16a for DFED-11E, excitation in the primary absorption band of the dye moiety at 15 K produces two paramagnetic species, assigned here as [Cl5Os(NO)]3-‚3V and D•+-XY. Annealing in the dark to 200 K for 10 min and quenching back to 15 K caused the conversion of [Cl5Os(NO)]3-‚3V to its twovacancy analogue, as indicated by the change in g factors shown in the figure. However, this annealing did not affect D•+-XY. In fact, double integration of the two spectra showed that the total concentration of radicals remained constant during the annealing treatment. The same series of experiments performed on grains sensitized with DFED-11 showed that the oxidized species D•+XY was destabilized by the absence of the blocking ester groupssee Figure 16b. No new types of radicals were detected as a result of its decay, but integration revealed, again, that annealing at 200 K did not change the total radical concentration in the sample. Thus, the loss of D•+-XY radicals caused by annealing must be compensated for by an increase in the concentration of reduced dopant traps. This can only be explained if the decay of the oxidized sensitizer results in electron injection into the silver halide substrate, consistent with the mechanism proposed in Scheme 3. Ab Initio Calculations of FED Molecules. We calculated molecular properties for PMT-3 and PMT-41 using ab initio methods. First, optimized molecular structures were calculated for their diamagnetic ground states in the gas phase using the

quantum-mechanical package Gaussian 0317 and the MidiX basis set. Hole trapping by the FED molecule was simulated by removing one electron. The molecular structure was subsequently allowed to relax. In this second optimization process, the CO2 group fragmented from the FED molecule after a few optimization steps. The initial and final geometries calculated for PMT-3•+ are shown in Figure 17a,b. All of the atoms were allowed to relax during the optimization process without restrictions. The result in Figure 17b predicts the loss of CO2. A similar fragmentation was also predicted for PMT-41•+. In addition to this theoretical verification of the fragmentation process, the ab initio calculations allow us to calculate spin densities for the FED radicals before and after fragmentation. For the PMT-3•+ and PMT-41•+ radicals before fragmentation, the spin density is distributed over the CO2 moiety (25%), the S atom (33%), and the central N atom (13%). The largest 1H Fermi contact interactions are calculated at 11 MHz for PMT3•+ and 15 MHz for PMT-41•+. After fragmentation, the spin density is largely confined to the C atom that was bonded to the CO2 group (77%). The largest 1H Fermi contact interaction is calculated at 78 MHz for the PMT-3-derived radical and 67 MHz for its PMT-41 analogue. 1H hyperfine interactions were detected with ENDOR for the PMT-41 radical (Figure 6). The exposure temperature of 100 K was below the onset temperature for the increase in Rh2+ signal and is therefore considered to be below the thermal threshold for the fragmentation process. The hyperfine interaction observed in ENDOR was 6 MHz, which is reasonably close to the calculated value of 15 MHz for PMT-41•+, considering that the influence of the AgBr substrate or the peptizing medium are not taken into account. In EPR, we observed potential 1H hyperfine splittings in a dispersion containing PMT-3 after band gap exposure at 15 K and subsequent annealing to 160 K (Figure 8). We have argued that this annealing temperature is high enough for the PMT3•+ fragmentation to occur. The experimentally observed splittings of about 2 mT correspond to a hyperfine interaction of 56 MHz. This is considerably larger than the 11 MHz calculated for PMT-3•+ but reasonably close to the maximum 1H hyperfine interaction calculated for the fragmented PMT-3 radical in its optimized gas-phase geometry (78 MHz). Summary and Conclusions In these experiments, we have used EPR spectroscopy to study the processes of hole capture, donor fragmentation, and



Figure 17. (a) The initial gas-phase geometry calculated for the PMT-3•+ radical produced by hole capture. (b) The optimized gas-phase geometry predicted for this FED radical showing decarboxylation is the preferred stabilization mechanism.

secondary electron injection in AgBr dispersions sensitized with fragmentable electron donors (FEDs). To do so, we employed AgBr grains doped with diamagnetic transition metal complexes that act as deep electron traps. We were then able to observe the concentration of both trapped electrons and trapped holes after low-temperature exposure of the dispersions and subsequent annealing to various temperatures. The EPR data presented here support the conclusion that the proposed two-electron sensitization mechanism for FED compounds, previously verified in solution, takes place at the AgX surface and provides extra electrons in the conduction band of AgX. For the FED compound PMT-41, we observe very strong hole trapping by the FED moiety and an actual increase in the number of electrons at Rh centers as the dispersion is annealed to room temperature. For PMT-3, we also see strong hole trapping and suppression of the loss of Rh2+ centers as the dispersion is annealed to room temperature. For PMT-3, the radical becomes unstable at a significantly lower temperature than for PMT-41, in agreement with PMT-3’s faster estimated room-temperature fragmentation rate. This lower temperature of fragmentation allows the observation of the decarboxylated radical by EPR. Assignments of the observed radicals are supported by our ab initio calculations of the expected hyperfine splitting constants. Experiments with PMT-3 and the blue-absorbing cyanine dye BTHC show that the two-electron sensitization mechanism also operates as predicted in the presence of a sensitizing dye. This conclusion is further supported by experiments with DFED11, where the FED moiety is directly linked to the sensitizing dye. The presence of the reference mercaptotetrazole a-PMT in the AgBr dispersions is found to modify the intrinsic [(Br4)3-‚V] hole center. This new center appears to be more thermally stable than the original [(Br4)3-‚V] center. Consequently, the sites where this modified center forms compete with the blue sensitizing dye for surface holes released from the original intrinsic hole centers during the thermal annealing process. Consistent with a-PMT’s inability to fragment and inject electrons, we observe no increases in the trapped electron concentration in these a-PMT-containing dispersions as they are annealed to room temperature. Experimental Details Materials. For this study, dispersions of cubic silver bromide microcrystals (or grains) were precipitated in aqueous gelatin

by standard double-jet techniques.18 Crystal edge lengths were typically 0.10 ( 0.01 µm. The grains were doped with the diamagnetic anions [(H2O)RhCl5]2- and [Cl5Os(NO)]2- added as aqueous solutions during the precipitation of the AgBr grains. In each case, the nominal dopant concentration was 100 molar parts per million (mppm). Dyed grains were obtained by adding a methanol solution of the dyes BTHC, DFED-11, or DFED11E to the melted dispersion at 40 °C and holding for 20 min to ensure dye adsorption. Dyes were added at 2.0 mmol/mol of Ag, an amount sufficient to cover approximately one-half the available surface area of the grains. For adsorption of the PMT compounds to the grain surface, aqueous solutions of a-PMT, PMT-3, or PMT-41 were added to the melted dispersion, which was then held for 5 min and chill set. For samples containing both BTHC dye and a PMT compound, the dye was added to the melted dispersion first. To obtain samples suitable for EPR and ENDOR studies, the dispersions were evaporated to dryness in a flowing N2 stream. The resultant dispersions were about 1 mm thick and opaque. They were handled in the dark and stored at 4 °C. Estimation of Solution Oxidation Potentials and Fragmentation Rates. Oxidation potentials were measured by cyclic or AC voltammetry, using a Princeton Applied Research Corp. (PAR) 173 potentiostat in conjunction with a PAR 175 universal programmer and a PAR 179 digital coulometer. Solutions for voltammetric analysis contained acetonitrile, 0.1 M tetrabutylammonium fluoroborate, and ∼5 × 10-4 M analyte. All solutions were deaerated with argon prior to the measurement. The working electrode was a Pt disk (∼0.02 cm2) that was polished with 0.1 µm diamond paste (Buhler Metadi) and rinsed with water before each experiment. All potentials were measured vs the NaCl saturated calomel electrode at 22 °C and reported vs an Ag/AgCl electrode. Measurements of the formal oxidation potential of the parent electron donor X-Y are complicated by the fast fragmentation reactions of the radical cations X-Y•+, which lead to significant shifts in the peak potentials to values lower than the formal potentials. Consequently, estimates of the XY oxidation potentials are best obtained from measurements of structurally related N,N-dimethylaniline model compounds. Formal oxidation potentials measured in acetonitrile yield values of 0.77, 0.91, and 1.18 V for the model anilines where the substituent R1 in the structure in Scheme 2 is equal to methyl, hydrogen, and formyl, respectively. The oxidation potential values for the specific FED

10136 J. Phys. Chem. B, Vol. 109, No. 20, 2005 compounds studied here are expected to be within 50 mV of the related model compound. The fragmentation rates for the XY compounds are also sensitive to the substituent R1. When R3 is methyl, R2 is acetate, and R1 is methyl, flash photolysis measurements19 for XY in acetonitrile/water indicate the fragmentation rate to be 8.1 × 106 s-1. For the analogous XY where R1 is hydrogen, similar measurements indicate the fragmentation rate to be slightly greater than 1 × 108 s-1. Unfortunately, the fragmentation rate for the corresponding formyl-substituted XY is much too fast to be directly measurable. However, it is found that the fragmentation rates for these compounds roughly correlate with oxidation potential,20 suggesting a rate constant for the formyl derivative approaching 1011 s-1. With use of this information on the fragmentation rates of these model compounds, the fragmentation rate rough estimates for the FED moieties in DFED-11, PMT-41, and PMT-3 become 107, 108, and 1011 s-1, respectively. To obtain an oxidation potential estimate for the compound a-PMT, a measurement was made on an analogue where the S was chemically blocked to mimic bonding to the AgX surface and to prevent adsorption of the analyte to the electrode. The value obtained for the irreversible oxidation potential was 1.9 V, suggesting a reversible potential of approximately 2.1 V. EPR Measurements. EPR spectra at 9.3 GHz were obtained with a Bruker ESP 300E spectrometer fitted with a helium gasflow cryostat. In situ irradiations were performed with the output of a 200 W super pressure Hg/Xe light source passing through a combination of a 1/8 m monochromator and interference and long-pass filters. EPR measurements at 35 GHz were made with a Varian E-12 spectrometer fitted with a helium gas-flow cryostat. ENDOR measurements were performed on a Bruker ER200D X-band EPR/ENDOR spectrometer. Ab Initio Calculations. The quantum mechanical procedure employed was the density functional method with the B3LYP functional, using the MidiX basis set. This method is implemented in the Gaussian 03 software package. All calculations were performed in the gas phase. The geometry of the groundstate XY was first optimized. Then the unpaired electron distribution was calculated for the XY•+ radical. A second optimization led to the fragmentation of the radical into X• + Y+ and produced the new unpaired electron distribution of X•. Acknowledgment. We thank Al Marchetti and Jeri Mount for providing the doped AgBr dispersions; Don Crosby, Bonnie Hein, and Renee Daniels for sample preparation; Chin H. Chen, Steve Godleski, and Paul Zielinski for providing the FED compounds; and Samir Farid and Ian Gould for useful discussions.

Eachus et al. References and Notes (1) Gould, I. R.; Lenhard, J. R.; Muenter, A. A.; Godleski, S. A.; Farid, S. Y. J. Am. Chem. Soc. 2000, 122, 11934. (2) Farid, S. Y.; Lenhard, J. R.; Chen, C. H.; Muenter, A. A.; Gould, I. R.; Godleski, S. A.; Zielinski, P. A. U.S. Patent 5,747,235, 1998. (3) Farid, S. Y.; Lenhard, J. R.; Chen, C. H.; Muenter, A. A.; Gould, I. R.; Godleski, S. A.; Zielinski, P. A.; Weidner, C. H. U.S. Patent 5,747,236, 1998. (4) Farid, S. Y.; Lenhard, J. R.; Chen, C. H.; Muenter, A. A.; Gould, I. R.; Godleski, S. A.; Zielinski, P. A.; Weidner C. H. U.S. Patent 5,994, 051, 1999. (5) Farid, S. Y.; Lenhard, J. R.; Chen, C. H.; Muenter, A. A.; Gould, I. R.; Godleski, S. A.; Zielinski, P. A. U.S. Patent 6,010,841, 2000. (6) West, W.; Gilman, P. B. In The Theory of the Photographic Process, 4th ed.; James, T. H., Ed.; MacMillan: New York, 1977; Chapter 10. (7) Marchetti, A. P.; Muenter, A. A.; Baetzold, R. C.; McCleary, R. T. J. Phys. Chem. B 1998, 102, 5287. (8) Pawlik, Th. D.; Eachus, R. S.; McDugle, W. G.; Baetzold, R. C. J. Phys.: Condens. Matter 1998, 10, 11795. (9) Eachus, R. S.; Baetzold, R. C.; Pawlik, Th. D.; Poluektov, O. G.; Schmidt, J. Phys. ReV. B 1999, 59, 8560. (10) Eachus, R. S.; Pawlik, Th. D.; Baetzold, R. C. J. Phys.: Condens. Matter 2000, 12, 8893. (11) Tani, T. Chemical Sensitization and Stabilization. In Photographic SensitiVity: Theory and Mechanisms; Oxford University Press: New York, 1995; pp 185-186. (12) (a) Eachus, R. S.; Marchetti, A. P.; Muenter, A. A. Annu. ReV. Phys. Chem. 1999, 50, 117. (b) Marchetti A. P.; Burberry, M. Phys. ReV. B 1988, 37, 10862. (13) Herz, A. H. AdV. Colloid Interface Sci. 1977, 8, 237. (14) Lenhard, J. J. Imaging Sci. 1986, 30, 27. (15) Pawlik, Th. D.; Eachus, R. S.; Baetzold, R. C.; McDugle, W. G. J. Phys.: Condens. Matter 2000, 12, 2555. (16) Barclay, T. M.; Hicks, R. G.; Ichimura, A. S.; Patenaude, G. W. Can. J. Chem./ReV. Can. Chim. 2002, 80 (11), 1501. (17) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Montgomery, J. A., Jr.; Vreven, T.; Kudin, K. N.; Burant, J. C.; Millam, J. M.; Iyengar, S. S.; Tomasi, J.; Barone, V.; Mennucci, B.; Cossi, M.; Scalmani, G.; Rega, N.; Petersson, G. A.; Nakatsuji, H.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Klene, M.; Li, X.; Knox, J. E.; Hratchian, H. P.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Ayala, P. Y.; Morokuma, K.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Zakrzewski, V. G.; Dapprich, S.; Daniels, A. D.; Strain, M. C.; Farkas, O.; Malick, D. K.; Rabuck, A. D.; Raghavachari, K.; Foresman, J. B.; Ortiz, J. V.; Cui, Q.; Baboul, A. G.; Clifford, S.; Cioslowski, J.; Stefanov, B. B.; Liu, G.; Liashenko, A.; Piskorz, P.; Komaromi, I.; Martin, R. L.; Fox, D. J.; Keith, T.; Al-Laham, M. A.; Peng, C. Y.; Nanayakkara, A.; Challacombe, M.; Gill, P. M. W.; Johnson, B.; Chen, W.; Wong, M. W.; Gonzalez, C.; Pople, J. A. Gaussian 03, Revision B.05; Gaussian, Inc., Wallingford, CT, 2004. (18) Berry C. R. In The Theory of the Photographic Process, 4th ed.; James, T. H., Ed.; MacMillan: New York, 1977; Chapter 3. (19) Gould, I. R.; Farid, S. Y. Private communication. (20) Gould, I. R.; Lenhard, J. R.; Farid, S. J. Phys. Chem. A 2004, 108, 10949.

An EPR Study of Two-Electron Sensitization by Fragmentable Electron

Recommend Documents