Laser-Induced Fluorescence of Rhodamine 6G Cations in the Gas

J. Phys. Chem. A 2010, 114, 5509–5514

5509

Laser-Induced Fluorescence of Rhodamine 6G Cations in the Gas Phase: A Lower Bound to the Lifetime of the First Triplet State Mattias Kordel,†,§ Detlef Schooss,*,†,‡ Christian Neiss,†,| Lars Walter,† and Manfred M. Kappes*,†,‡ Institut fu¨r Nanotechnologie, Karlsruher Institut fu¨r Technologie, Postfach 3640, 76021 Karlsruhe, Germany, and Institut fu¨r Physikalische Chemie, Karlsruher Institut fu¨r Technologie, Kaiserstrasse 12, 76128 Karlsruhe, Germany ReceiVed: January 22, 2010; ReVised Manuscript ReceiVed: March 24, 2010

We have studied the gas-phase laser-induced fluorescence of an ensemble of buffer gas-cooled Rhodamine 6G cations (R6G+) stored in a quadrupole ion trap at 90 K. The fluorescence resulting from excitation with continuous-wave 488 nm radiation was observed to disappear almost completely on a time scale of seconds, dependent in detail on the excitation laser fluence. Such decay can be explained by the accumulation of R6G+ in a dark triplet state. This in turn facilitates the first lifetime determination of the lowest triplet state of free R6G+ by direct ground-state recovery measurements. A lower bound for the half-life was found to be ∼2 s. Adding oxygen in a volume fraction of 1% to the buffer gas leads to efficient quenching of the triplet state and correspondingly to complete suppression of the fluorescence intensity decay. Different rare gases were applied as buffers for collisional cooling, but no significant changes in the fluorescence properties were found. SCHEME 1: Structure of Rhodamin 6G+

I. Introduction Rhodamine dyes have superior photochemical and photophysical properties such as high fluorescence quantum yields and photostability. These properties have given rise to widespread technological and scientific uses, for example, as laser dyes.1 A particularly well-studied member of the Rhodamine family of xanthene dyes is Rhodamine 6G. This fluorophore is typically obtained as the monochloride salt (R6G-Cl; C28H31N2O3Cl; scheme 1) and as such is well soluble in many polar solvents. The photophysics of R6G+ in liquid solution or embedded in various transparent solid matrices is very well established. Examples of frequency- and time-domain spectroscopic studies that give insight into state energies, transition cross sections, and interconversion rates are given in refs 2 and 3. The field of few or single molecule spectroscopy has also made extensive use of R6G+ as a model fluorophore. Triplet state dynamics of solvated R6G+ molecules has been studied by fluorescence correlation spectroscopy (FCS).4 The associated photoblinking off times have been shown to be sensitive to the presence of oxygen.5 Two-step photobleaching by way of an intermediate triplet state has also been studied by FCS.6 More recently, photobleaching of R6G+ in polyvinyl-alcohol matrix was probed by Orrit et al.7,8 Beyond triplet R6G+, a second type of dark intermediate state was postulated: long-lived radical anions whose formation mechanism remains unclear. In these studies, oxygen is generally considered to be the most important photobleaching reagent for R6G+ at room temperature. The corresponding photooxidation is believed to involve singlet * To whom correspondence should be addressed. E-mail: detlef.schooss@ kit.edu, [email protected]. † Institut fu¨r Nanotechnologie, Karlsruher Institut fu¨r Technologie. ‡ Institut fu¨r Physikalische Chemie, Karlsruher Institut fu¨r Technologie. § Present address: Gleiss & Grosse, Leitzstrasse 45, D-70469 Stuttgart. | Present address: Lehrstuhl fu¨r Theoretische Chemie, Universita¨t Erlangen-Nu¨rnberg, Egerlandstr. 3, 91058 Erlangen, Germany.

(di)oxygen, itself generated by reaction of ground-state triplet dioxygen with the R6G+ triplet state. In condensed phase studies of single R6G+ fluorophores, many open questions revolve around the influence of matrix/ solvent effects on the elementary photochemical processes. Such issues are typically folded with the wide range of different local environments present. As a result, the distinction between elementary processes that are primarily molecule-intrinsic (e.g., intersystem crossing) and those that involve environmental coupling (e.g., triplet quenching or constrained diffusion of singlet oxygen photoproducts followed by oxidative bleaching) can sometimes be unclear. Laser-induced fluorescence (LIF) in the gas phase is capable of yielding information on bare atomic or molecular ions without any solvent or surface interactions, thus shedding light on the intrinsic properties of the system under study. In early measurements, molecular beam techniques were used to probe the fluorescence of atomic ions.9,10 Later, the method was extended to molecular ions (ref 11 and references therein). Parallel to this development, the first fluorescence measurements on trapped atomic ions were carried out12-15 and extended to small molecular ions.16-19 In this study, we have used trapped ion laser-induced fluorescence (TLIF) to study the photophysics of isolated R6G+ cations. The TLIF method has the advantage of providing luminescence emission data for the same ensemble of molecular ions

10.1021/jp100636x  2010 American Chemical Society Published on Web 04/14/2010

5510

J. Phys. Chem. A, Vol. 114, No. 17, 2010

over a long time period (i.e., over many absorption emission cycles). Ideally, this acquisition period is only limited by the mean trapping time, which can be on the order of seconds or even longer. Marshall and coworkers were the first to apply the method to larger molecular ions. They have determined spectral as well as time-dependent integral fluorescence for ensembles of C6F6+ and C6H3F3+ stored in a room temperature Penning trap.19-21 Parks et al. have employed a custom-built quadrupole ion trap to show that nearly background-free detection of the fluorescence of trapped organic dye ions (including Rhodamine 640 cations) is possible under pulsed laser excitation conditions.22 In addition, they have extensively investigated the motional (and folding) dynamics of dye-labeled biomolecules by spectrally dispersed fluorescence measurements using continuous-wave (cw) laser excitation at various trap temperatures.23-26 Zenobi et al. have studied the fluorescence of a number of different Rhodamine dye cations (including R6G+) stored within a Penning trap at room temperature.27 A strong increase in R6G+ fluorescence intensity was noted upon raising the helium buffer gas pressure. This was attributed (in part) to effective collisional cooling of photoexcited molecules and a corresponding reduction in the competing photofragmentation rate. Interestingly, complete quenching of R6G+ fluorescence emission was observed when switching to Ar buffer gas and irradiating the same ensemble of ions for an extended period of time.28 More recently, the Zenobi group has also probed the laser power and helium buffer gas pressure dependence of R6G+ emission in more detail. Laser excitation at 488 nm was found to cause photofragmentation to a cationic dissociation product, tentatively assigned as Rhodamine 575+, which also fluoresces. Ervin et al. have recently reported a series of TLIF measurements on Rhodamine 575 cations (and their photofragments) using a room-temperature quadrupole ion trap attached to a timeof-flight mass spectrometer.29,30 In this study, photofragmentation yield and integral fluorescence (parents + fragments) as excited by a cw laser were determined as functions of laser power, buffer gas pressure, and irradiation time. The data were qualitatively discussed in terms of underlying elementary processes, among them intersystem crossing from the photoexcited singlet to an intermediate dark state (which can be photodissociated in a second absorption step). In the present study, we present a quantitative study of the photophysics of an ensemble of R6G+ ions confined in a radio frequency (rf) quadrupole ion trap at 90 K and subject to cw laser irradiation over time periods of up to several seconds. We observe a time-dependent intensity decay when monitoring the integral fluorescence and explain this result with a simple threestate model, connecting two singlet states and a dark triplet state. We determine the half-life time of the triplet state directly by measuring the recovery rate from the dark triplet state to the singlet ground state. Our observations concerning integral fluorescence emission are found to be essentially independent of the nature of the buffer gas (at equal pressures of He, Ne, and Ar). In contrast, we observe very efficient R6G+ triplet quenching and corresponding recovery of fluorescence by collisions with dioxygen. II. Experimental Methods All experiments were carried out in a newly constructed TLIF apparatus, a schematic of which is shown in Figure 1. It consists of an electrospray ion source, ion optics, and a quadrupole ion trap coupled to a fluourescence microscope. Ion beams of R6G+ cations were generated by electrospray from a 10-5 M solution of R6G-Cl (chloride salt, Sigma-Aldrich) in methanol using a

Kordel et al.

Figure 1. Schematic of the trapped ion laser-induced fluorescence (TLIF) setup.

home-built electrospray ionization (ESI) source. The R6G+ ion beam was electrostatically steered and focused into an rf quadrupole ion trap, which was also built in house. The ion trap has a hyperbolic profile comprising an inner ring of diameter 14.14 mm and an end-cap spacing of 11.06 mm, that is, it corresponds to a stretched quadrupole trap configuration. The trap operates at a radiofrequency of 300 kHz with a peak-topeak amplitude of up to 7000 V. The RF electronics were also home-built and based on a design by Parks et al.31 The ion trap not only serves as an ion storage device during the fluorescence measurements but also as a mass spectrometer for ion isolation and detection. To isolate the desired mass-to-charge ratio, the method of stored waveform inverse Fourier transform (SWIFT)32,33 was used. For this, the appropriate waveform was applied dipolar to the end-caps of the ion trap. The ion trap walls can be regulated to temperatures between 90 and 650 K. For the experiments described here, the trap temperature was set to ∼90K. To trap ions effectively from the incoming ion beam (Ekin ) 10-20 eV), a buffer gas at a pressure of ca. 10-3 mbar was present in the trap. In the work presented here, we used He (g99.9999%), Ne (g99.999%), or Ar (g99.9999%) as buffer gases. Trapped R6G+ ions were excited by the 488 nm line of a cw Ar ion laser (Spectra-Physics, model 2080-15S) incident normal to the ion beam axis. During laser excitation, the buffer gas pressure within the trap was raised to and maintained at a comparatively high value (0.02 to 0.2 mbar).47 This ensures that any residual energy deposited into the ions by consecutive absorption and fluorescence cycles can be transferred to the buffer gas via collisions. The buffer gas was introduced to the trap by way of a tube attached to a hole in one of the ceramic spacers between ring and end-cap electrodes. The buffer gas flow was controlled by a fast solenoid valve (Parker Series 99), which was itself connected to a pressure-regulated reservoir. The pressure inside the ion trap was calculated from the steadystate pressure in the vacuum chamber with the valve open, as determined by a Bayard-Alpert gauge, under the assumption that the integral molecular flow through all trap orifices determines the overall buffer gas conductance. The power of the incident laser beam was attenuated in discrete steps by means of a set of neutral density filters. To clean up the incoming laser beam spectrally, a narrow bandpass filter (Semrock, MaxLine LL01-488-12.5) was used. A laser shutter provided for synchronization with the experiment. The laser beam was focused into the center of the quadrupole trap

LIF of Rhodamine 6G Cations in the Gas Phase using an externally mounted lens (f ) 500 mm) and passed through 1.2 mm diameter holes on opposite sides of the ring electrode. To suppress scattered light, the laser beam path contained a Brewster window and a set of baffle stacks, which were mounted within the vacuum chamber on both sides of the quadrupole trap. Finally, the transmitted laser beam was dumped onto a photo diode for power measurement. The laser beam diameter at trap center was measured to be ∼0.2 mm. Ion fluorescence was collected by an infinity-corrected microscope objective (NA ) 0.15, Zeiss, EC Plan-Neofluar 5×/ 0.15) through a 3 mm diameter aperture in the quadrupole trap end-cap (i.e., perpendicular to the excitation beam). The parallelized light beam emanating from the microscope objective passed through the vacuum chamber before exiting through a quartz window. Subsequently, the emitted light was passed through a laser edge filter (Semrock, RazorEdge LP02-488RS25) to remove scattered excitation laser light and focused onto the aperture of a photomultiplier tube (PMT, Hamamatsu, model H7421-40). The latter was read out with a counter card PXI6602 (National Instruments) having a temporal resolution of 100 kHz. The spatial extent of the ion cloud was measured by fluorescence ion tomography.34 At the typical ion numbers used in this work (∼2 × 104), the cloud is axially and also radially well described by a Gaussian density profile with diameters of 0.30 to 0.5 mm, depending on the buffer gas. This leads to a typical overlap between ion cloud and laser beam of ∼10%. A typical experimental cycle started with ion collection for 1 to 2 s by trapping from the incident ion beam. This was followed by a period during which a SWIFT waveform was applied to the trapped ion cloud, leading to 103-105 massselected R6G+ ions. The ion density in the center of the trap was then increased by raising the rf amplitude corresponding to a quadrupole trap parameter35 of qz ≈ 0.8. After a subsequent cooling period of up to 2 sec, the laser shutter was opened, and fluorescence was detected for a preset time. Both the cooling and irradiation steps were carried out at high buffer gas pressure. Finally, the buffer gas is pumped away, and a mass selective instability scan35 was applied to drive the ions out of the trap, where they were detected with a dynode/channeltron assembly. The whole sequence was then repeated without ions in the trap to obtain a reference measurement. III. Results and Discussion Upon cw irradiating an R6G+ ion cloud, as described in the previous section, strong fluorescence was observed. The corresponding fluorescence intensity decreases with time depending in detail on the incident laser power and the buffer gas pressure. Figure 1 shows typical intensity-time traces as monitored over 1 sec at different laser powers. As the laser power increases, the decay of the fluorescence clearly becomes faster. For long irradiation times, the fluorescence was observed to decay to a stationary nonzero level. (Specific values depend on laser intensity.) Note that during the fluorescence experiments shown, care was taken to ensure that the number of R6G+ ions was conserved. For this, the pressure of collision gas and the laser intensity were chosen such that neither charged photodissociation products of R6G+ nor parent ion losses were detectable. In addition, the number of ions without laser irradiation (but with collisions) was also observed to be constant over the experimental time scales used within experimental accuracy. We found that for all laser powers shown, the initial fluorescence intensity could be fully recovered by blocking the excitation beam for a sufficiently long time (see below).

J. Phys. Chem. A, Vol. 114, No. 17, 2010 5511

Figure 2. Fluorescence intensity (in arbitrary units) traces of R6G+ at 488 nm. The number of trapped ions was approximately 1.8 × 104. The laser power was varied from 0.8 to 168 W/cm2, and the He buffer gas pressure was 0.2 mbar. The red lines are fits of eq 4 to the data.

Figure 3. Three state model used to describe the fluorescence intensity of trapped R6G+. The energies given to the right of the states result from TDDFT calculations with the functional BP86 and a SVP basis.

To explain these observations, we next turn to a kinetic model. Figure 3 shows a simple Jablonski type diagram comprising relevant electronic states and possible transitions between them. The energies shown to the right of the corresponding levels result from time-dependent density functional theory (TDDFT) calculations, as carried out using the program package TURBOMOLE36,37 with the BP86 functional and an SVP basis set. S0 denotes the singlet ground state of R6G+, S1 is the first excited singlet state, and T1 is the lowest triplet state. We assume that vibrational relaxation within the excited electronic state is much faster than that of other transitions38 and can therefore be neglected in our simple kinetic model. Furthermore, the irradiation intensities used (max 168 W/cm2) were small enough such that excitation from S1 to higher singlet states can be neglected as well. We also consider that stimulated emission,39 internal conversion (S1 to S0),40,41 as well as triplet-triplet absorption3 are insignificant under our conditions. The rate of S1 excitation may then be written as σP, where σ is the R6G+ absorption cross section and P is the laser intensity. In equations 1-3, we formulate (σP)eff as an effective excitation rate to take into account corrections for imperfect overlap of ion cloud and laser beam (as well as a dependence of this overlap on the number of trapped ions).42 Within our model, de-excitation from S1 is possible either via fluorescence (with a rate kfl) or by intersystem crossing (kISC) to the triplet state. Correspondingly, if the triplet state has a lifetime comparable to the experimental time scale, then ions are accumulated in the triplet state and removed from the excitation-fluorescence cycle; they become “dark”. As the laser power is increased,

5512

J. Phys. Chem. A, Vol. 114, No. 17, 2010

Kordel et al.

TABLE 1: Rate Parameters (σP)eff and kT Obtained from the Fits of the Data in Figure 1a P/W cm-2

(σP)eff/s-1

kT/s-1

0.8 4.3 43 88 168

78 415 3940 7700 15800

0.5 0.38 0.42 0.51 0.76

Intersystem crossing rate was held fixed at kISC ) 106 s-1, the fluorescence de-excitation rate at kfl ) 2.5 × 108 s-1. a

one would then expect a larger relative concentration of R6G+ molecules to be converted to the triplet state within a shorter time period. The differential equations governing the kinetics of these processes can be formulated as

d[S0] ) -(σP)eff[S0] + kfl[S1] + kT[T1] dt

(1)

d[S1] ) (σP)eff[S0] - (kfl + kISC)[S1] dt

(2)

d[T1] ) kISC[S1] - kT[T1] dt

(3)

We assume that the fluorescence intensity, I(t), is proportional to the number of R6G+ in the singlet excited state [S1]. Assuming kfl . kISC, kT, an approximate solution for the fluorescence intensity as a function of irradiation time, is given by4,41

I(t) ∝ 1+

{(

)

}

(σP)effkISC (σP)effkISC (t - t0) exp - kT + kT(kfl + (σP)eff) kfl + (σP)eff (σP)effkISC 1+ kT(kfl + (σP)eff) (4)

where t0 is an offset accounting for the time delay between triggering the PMT readout and opening the laser shutter. The red curves in Figure 2 are fits of expression 4 to the experimental intensity traces. The excellent quality (coefficients of determination >97%, except for 0.8 mW/cm2 trace) of these fits shows that this simple three-state model is sufficient to describe the observed behavior. Note that to obtain the above best fits, only (σP)eff, t0, and kT were independently varied, whereas kfl and kISC were fixed at constant values, as adopted from solution measurements. (See footnote in Table 1.)8 The use of these values is reasonable because condensed phase measurements have established that fluorescence and intersystem crossing rates of R6G+ are largely independent of external parameters.7 In our kinetic model, the only rate that is directly accessible experimentally is the recovery rate from the triplet state, kT. The inset of Figure 4 shows the experimental scheme, which we have applied to determine kT independent of the other rates. First, the ions are pumped to a stable triplet concentration via long time-scale irradiation. Then, the laser is shut off, and the ions are stored in the dark for a variable delay time, τ. After this delay, the laser shutter is reopened, and the measurement of the fluorescence intensity time trace is repeated for the same

Figure 4. Direct measure of the triplet to singlet ground-state recovery rate, kT, using the scheme shown in the inset. Pure He (0.027 mbar) was used as a buffer gas at a trap temperature of 90 K and an excitation intensity of 124 W/cm2. The exponential fit yields a rate of kt ) 0.48 ( 0.02 s-1.

ion cloud. During the “laser-off” delay, R6G+ ions can return from the triplet state to the singlet ground state, whereas no additional ions can be converted to the triplet state. Correspondingly, the fluorescence intensity at the beginning of the second measurement depends on the delay time, τ. If one assumes that the (relative) concentration in the S1 state is very small at all delay times, then the sum of the relative concentrations of S0 and T1 states is approximately unity. Correspondingly, the relative number of ions in the triplet state can be written as

[T1](τ) ) 1 -

I(τ) I0

(5)

where I(τ) is the fluorescence intensity at the beginning of the second measurement, as carried out after a delay time, τ, and I0 is the intensity at the beginning of the first measurement (before preparation of any triplet state species). Figure 4 shows a typical result of such an experiment. The data is consistent with a single exponential decay, as described by a rate kT. The fit yields kT ) 0.48 ( 0.02 s-1 corresponding to a triplet state lifetime on the order of 2 s. We note that this lifetime should be regarded as a lower bound to the lifetime of the triplet state of free R6G+ because contributions from as yet unknown quench processes cannot be fully excluded. Additional support for the suggested mechanism results from fluorescence measurements using buffer gas, which contains molecular oxygen. Figure 5 shows a fluorescence intensity time trace measured using a laser intensity of 168 W/cm2 and 55 × 10-3 mbar of a He mixture containing 1 vol % oxygen. The observed fluorescence is now found to be essentially timeindependent, even for the highest excitation intensities used in this study. This finding supports the inference of increasing conversion to a long-lived “dark” triplet state as the reason for the observed fluorescence decay when using oxygen-free inert buffer gas. Obviously, the recovery rate from the triplet state, kT, is considerably increased by the presence of molecular oxygen. The efficiency of dioxygen (X3Σg-) as a triplet quencher of R6G+ in the gas phase is apparently quite high. A simulation of the data based on eq 4 shows that the effective triplet recovery rate must be (kT)eff > 104 s-1 for there to be no detectable decay in the corresponding fluorescence trace. Given that the ions in the trap undergo ∼2 × 104 collisions per second with oxygen molecules, at least every other collision must be active for triplet quenching.

LIF of Rhodamine 6G Cations in the Gas Phase

J. Phys. Chem. A, Vol. 114, No. 17, 2010 5513 Acknowledgment. We would like to thank Matthias Schmid for assistance with the measurements and Dr. Sergei Lebedkin and Dr. Jean Francois Greisch for helpful discussions. M.M.K. acknowledges support of this work by the Helmholtz-Gemeinschaft and by the Deutsche Forschungsgemeinschaft through the Center for Functional Nanostructures. References and Notes

Figure 5. Fluorescence intensity time trace for Rhodamine 6G cations using 55 × 10-3 mbar He with 1% oxygen as buffer gas. The excitation intensity was 168 W/cm2.

TABLE 2: Recovery Rates, kT, of R6G+ for Different Buffer Gases gas

pressure/10-3 mbar

kT/s-1

He Ne Ar

27 20 34

0.48 ( 0.01 0.33 ( 0.02 0.34 ( 0.01

Dashtiev et al.28 were unable to observe fluorescence emission from Penning-trapped R6G+ when using Ar instead of He as buffer gas. This was explained by an increased intersystemcrossing rate, kISC, as mediated by spin-orbit coupling. We have repeated our fluorescence experiments using Ne and Ar buffer gases in the same pressure range as previously used by us for He. We do not observe a significant reduction of the fluorescence intensity when using either Ne or Ar. In addition, the triplet recovery rates observed for Ne and Ar were found to be even smaller than the value seen for He but still within a factor of two of the latter. Table 2 summarizes our results. IV. Conclusions We have measured the integral LIF of ensembles of ca. 104 trapped R6G+ at 90 K and at an excitation wavelength of 488 nm. The buffer gas (He, Ne, or Ar) pressure inside the quadrupole ion trap was chosen such that collisonal cooling precludes any detectable dissociation of R6G+. Fluorescence intensity-time traces at various laser powers can be wellmodeled in terms of a three-level scheme, which includes a dark triplet state. Fluorescence recovery measurements indicate that under our conditions the lowest triplet state lifetime of R6G+ is >2 s-1. It is interesting to note that this is orders of magnitude longer than the accepted range of lowest triplet state lifetimes, as determined for R6G+ in various condensed phase environments by optical spectroscopy.7,43 In contrast, electron spin resonance probes of frozen ethanol solutions at 77 K support R6G+ triplet state lifetimes on the order of seconds.44,45 Groundstate dioxygen is known to be an effective triplet-quencher in condensed phase.46 Here we have shown that in the gas phase, triplet R6G+ is quenched upon essentially every collision with dioxygen. We have demonstrated that it is possible to generate a >99% relative concentration of (metastable) triplet R6G+ species via optical pumping in an ion trap. This opens the possibility for triplet reaction studies, for example, to examine photobleaching mechanisms. Dispersed fluorescence measurements to unravel the influence of solvent molecule complexation on the photophysics of R6G+ are currently also under way.

(1) Drexhage, K. H. Dye lasers. In Topics in Applied Physics; Scha¨fer, F. P., Ed.; Springer-Verlag: New York, 1973; Vol. 1. (2) Dempster, D. N.; Morrow, T.; Quinn, M. F. J. Photochem. 1973, 2, 343. (3) Korobov, V. E.; Shubin, V. V.; Chibisov, A. K. Chem. Phys. Lett. 1977, 45, 498. (4) Widengren, J.; Mets, U.; Rigler, R. J. Phys. Chem. 1995, 99, 13368. (5) Weston, K. D.; Carson, P. J.; DeAro, J. A.; Buratto, S. K. Chem. Phys. Lett. 1999, 308, 58. (6) Eggeling, C.; Widengren, J.; Rigler, R.; Seidel, C. A. M. Anal. Chem. 1998, 70, 2651. (7) Zondervan, R.; Kulzer, F.; Orlinskii, S. B.; Orrit, M. J. Phys. Chem. A 2003, 107, 6770. (8) Zondervan, R.; Kulzer, F.; Kol’chenko, M. A.; Orrit, M. J. Phys. Chem. A 2004, 108, 1657. (9) Andra¨, H. J.; Gaupp, A.; Wittmann, W. Phys. ReV. Lett. 1973, 31, 501. (10) Holt, R. A.; Rosner, S. D.; Gaily, T. D.; Adam, A. G. Phys. ReV. A 1980, 22, 1563. (11) Maier, J. P. Chem. Soc. ReV. 1988, 17, 45. (12) Bialas, J.; Blatt, R.; Neuhauser, W.; Toschek, P. E. Opt. Commun. 1986, 59, 27. (13) Blatt, R.; Zoller, P.; Holzmueller, G.; Siemers, I. Z. Phys. D 1986, 4, 121. (14) Bergquist, J. C.; Wineland, D. J.; Itano, W. M.; Hemmati, H.; Daniel, H.-U.; Leuchs, G. Phys. ReV. Lett. 1985, 55, 1567. (15) Plumelle, F.; Desaintfuscien, M.; Duchene, J. L.; Audoin, C. Opt. Commun. 1980, 34, 71. (16) Danon, J.; Mauclaire, G.; Govers, T. R.; Marx, R. J. Chem. Phys. 1982, 76, 1255. (17) Grieman, F. J.; Mahan, B. H.; O’Keefe, A. J. Chem. Phys. 1980, 72, 4246. (18) Martner, C. C.; Pfaff, J.; Rosenbaum, N. H.; O’Keefe, A.; Saykally, R. J. J. Chem. Phys. 1983, 78, 7073. (19) Wang, Y.; Hendrickson, C. L.; Marshall, A. G. Chem. Phys. Lett. 2001, 334, 69. (20) Cage, B.; Friedrich, J.; Little, R. B.; Wang, Y.-S.; McFarland, M. A.; Hendrickson, C. L.; Dalal, N.; Marshall, A. G. Chem. Phys. Lett. 2004, 394, 188. (21) Friedrich, J.; Fu, J.; Hendrickson, C. L.; Marshall, A. G. ReV. Sci. Instrum. 2004, 75, 4511. (22) Khoury, J. T.; Rodriguez-Cruz, S. E.; Parks, J. H. J. Am. Soc. Mass Spectrom. 2002, 13, 696. (23) Danell, A. S.; Parks, J. H. J. Am. Soc. Mass Spectrom. 2003, 14, 1330. (24) Danell, A. S.; Parks, J. H. Int. J. Mass Spectrom. 2003, 229, 35. (25) Iavarone, A. T.; Parks, J. H. J. Am. Chem. Soc. 2005, 127, 8606. (26) Iavarone, A. T.; Duft, D.; Parks, J. H. J. Phys. Chem. A 2006, 110, 12714. (27) Frankevich, V.; Guan, X.; Dashtiev, M.; Zenobi, R. Eur. J. Mass. Spectrom. 2005, 11, 475. (28) Dashtiev, M.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2006, 17, 855. (29) Sassin, N. A.; Everhart, S. C.; Dangi, B. B.; Ervin, K. M.; Cline, J. I. J. Am. Soc. Mass Spectrom. 2009, 20, 96. (30) Chingin, K.; Chen, H. W.; Gamez, G.; Zenobi, R. J. Am. Soc. Mass Spectrom. 2009, 20, 1731. (31) Parks, J. H.; Pollack, S.; Hill, W. J. Chem. Phys. 1994, 101, 6666. (32) Chen, L.; Wang, T.-C. L.; Ricca, T. L.; Marshall, A. G. Anal. Chem. 1987, 59, 449. (33) Marshall, A. G.; Wang, T.-C. L.; Ricca, T. L. J. Am. Chem. Soc. 1985, 107, 7893. (34) Kordel, M. Thesis, Universita¨t Karlsruhe, 2007. (35) March, R. E.; Todd, J. F. J. Practical Aspects of Ion Mass Spectrometry; CRC Press: Boca Raton, FL, 1995; Vol. 1. (36) Ahlrichs, R.; Ba¨r, M.; Ha¨ser, M.; Horn, H.; Ko¨mel, C. Chem. Phys. Lett. 1989, 162, 165. (37) Furche, F.; Ahlrichs, R. J. Chem. Phys. 2002, 117, 7433. (38) Penzkofer, A.; Falkenstein, W.; Kaiser, W. Chem. Phys. Lett. 1976, 44, 82. (39) Beaumont, P. C.; Johnson, D. G.; Parsons, B. J. J. Chem. Soc., Faraday Trans. 1993, 89, 4185.

5514

J. Phys. Chem. A, Vol. 114, No. 17, 2010

(40) Huth, B. G.; Kagan, M. R. IBM J. Res. DeV. 1971, 15, 278. (41) Zondervan, R.; Kunzler, F.; Orlinskii, S. B.; Orrit, M. J. Phys. Chem. A 2003, 107, 6770. (42) Kordel, M.; Kappes, M.; Schooss, D. ReV. Sci. Instrum., to be published. (43) Snavely, B. B.; Scha¨fer, F. P. Phys. Lett. A 1969, 28, 728. (44) Antonucci, F. R.; Tolley, L. G. J. Phys. Chem. 1973, 77, 2712.

Kordel et al. (45) Yamashita, M.; Kashiwagi, H. J. Phys. Chem. 1974, 78, 2006. (46) Hubner, C. G.; Renn, A.; Renge, I.; Wild, U. P. J. Chem. Phys. 2001, 115, 9619. (47) The corresponding collision rates are estimated to be 8 × 10-5 to 8 × 10-6 s-1 using a Langevin collosion rate coefficient and assuming a thermal ion distribution.

JP100636X