Emission Lifetimes of a Fluorescent Dye under Shock Compression

Oct 15, 2015 - We know of one previous study of emission lifetimes under shock. ...... A. D.; Banishev , A. A.; Shaw , W. L.; Dlott , D. D. Laser-Driv...
1 downloads 0 Views 832KB Size
Subscriber access provided by CMU Libraries - http://library.cmich.edu

Article

Emission Lifetimes of a Fluorescent Dye under Shock Compression Weilong Liu, Will P. Bassett, James M. Christensen, and Dana D. Dlott J. Phys. Chem. A, Just Accepted Manuscript • DOI: 10.1021/acs.jpca.5b09695 • Publication Date (Web): 15 Oct 2015 Downloaded from http://pubs.acs.org on October 19, 2015

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

The Journal of Physical Chemistry A is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Emission Lifetimes of a Fluorescent Dye under Shock Compression Wei-long Liu&, Will P. Bassett, James M. Christensen and Dana D. Dlott* School of Chemical Sciences and Fredrick Seitz Materials Research Laboratory, University of Illinois at Urbana-Champaign, Urbana, Il 61801, USA Abstract The emission lifetimes of rhodamine 6G (R6G), were measured under shock compression to 9.1 GPa, with the dual intent of better understanding molecular photophysics in extreme environments and assessing the usefulness of fluorescence lifetime microscopy to measure spatially-dependent pressure distributions in shocked microstructured media. R6G was studied as free dye dissolved in poly-methyl methacrylate (PMMA), or dye encapsulated in silica microparticles suspended in PMMA. Thin layers of these materials in impedance-matched geometries were subjected to planar single-stage shocks created by laser-driven flyer plates. A synchronized femtosecond laser excited the dye at selected times relative to flyer plate arrival and the emission lifetimes were measured with a streak camera. Lifetimes decreased when shocks arrived. The lifetime decrease was attributed to a shock-induced enhancement of R6G nonradiative relaxation. At least part of the relaxation involved shock-enhanced intersystem crossing. For free dye in PMMA, the lifetime decrease during the shock was shown to be a linear function of shock pressure from 0-9 GPa, with a slope of -0.22 ns·GPa-1. The linear relationship makes it simple to convert lifetimes into pressures. Lifetime measurements in shocked microenvironments may be better than emission intensity measurements, since lifetimes are sensitive to the surrounding environment, but insensitive to intensity variations associated with the motion and optical properties of a dynamically changing structure.

& *

Present address, Department of Physics, Harbin Institute of Technology, Harbin, 150001, China Author to whom correspondence should be addressed. Electronic mail [email protected]

1 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

keywords: nonradiative process, high pressure, shock waves

2 ACS Paragon Plus Environment

Page 2 of 29

Page 3 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

1. Introduction In this study we investigate how the emissive lifetime of a dye, rhodamine 6G (R6G), is affected by shock compression. The dual intents of our study are to learn more about molecular photophysics in extreme environments using a well-studied molecular emitter, and to evaluate the possibilities of using fluorescence lifetime microscopy measurements1 to study shocks in microstructured media.

In a shocked microstructured medium,2 the strains, pressures,

temperatures3 and compositions are complex functions of space and time.4 Such distributions are currently difficult to probe in real time. In our experiments, an impact with a laser-driven flyer plate created a planar shock in the 0-9 GPa range with 18 ns duration in a thin layer of dye-doped PMMA. A single femtosecond pulse was used to excite R6G at different times before, during and after the shock, and the emission was time resolved to determine the lifetime. We are not studying emission created by the shock on a ground state. We are studying how the emission from laser-generated excited states is affected by a shock. We studied R6G in two forms. Either R6G was dissolved in poly-methyl methacrylate (PMMA), or it was encapsulated in silica microparticles that were themselves suspended in PMMA.5

Silica-encapsulated dye particles have been described as “bright” and

“superfluorescent”,6,7 and we have shown that R6G microparticles emit 3.4 times more intensely than R6G dissolved in PMMA, termed “free dye”.5 Broadly speaking, there are three ways to use a fluorescent dye to probe shock compression dynamics: (1) the shock-induced spectral shift;8-10 (2) shock-induced changes of emission intensity;9,10 or (3) shock-induced changes of the emission lifetime. The shock effect

3 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 29

on the lifetime is due to photophysics, primarily changes in nonradiative rates due to environmental effects. We have previously studied shock-induced spectral shifts and intensity changes for the two forms of R6G in PMMA.5,11,12 Photophysics of R6G will be discussed in relation to the Jablonsky diagram11,13 in Fig. 1. Shocking R6G causes an emission redshift and an intensity loss.9-12 The redshift is a solvation effect. Dissolving R6G in PMMA causes a solvent redshift. PMMA lowers the energies of both S0 and S1, but the S1 lowering is greater since S1 is more polarizable.14,15 Compressing PMMA increases the solvation effect, lowering the S1 energy more than S0, producing a redshift. Shock-induced redshift is an effectively instantaneous probe of the density increase experienced by R6G. We know of one previous study of emission lifetimes under shock. Huston, Justus and Campillo found that the crystal violet lifetime was increased from its ambient value of 100 ps to 200 ps at 2 GPa and the increase was attributed to shock-induced changes in glycerol viscosity.16 The shock-induced redshift of R6G emission in ethanol solution up to 1.9 GPa was studied by Shen and Gupta in 1991,8 and later extended to 3.5 GPa by Ichiyanagi and co-workers.17 Both studies showed how the redshift could be used as an optical monitor of pressure.

The emission

lifetimes of R6G in ethanol were studied at static high pressures up to 0.9 GPa by Taguchi, Hirayama and Okamoto.18 The monomer lifetime was increased from its ambient 2.3 ns value to 3.6 ns at 0.45 GPa, at concentrations where aggregates were present and the medium was a fluid so R6G monomers and aggregates could diffuse toward and away from each other. The static high pressure lifetimes of a related dye, rhodamine B (RhB) in a solid matrix (poly-acrylic acid) were studied by Dreger and Drickamer.19 The RhB lifetime was decreased from 6 ns to 3 ns when pressure was increased from ambient to 7 GPa. The lifetime decrease was attributed to

4 ACS Paragon Plus Environment

Page 5 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

increased nonradiative rates. Dreger and Drickamer used pressure tuning arguments to suggest that internal conversion was more important than intersystem crossing.19 Although RhB is chemically similar to R6G, RhB emission is known to be more temperature-dependent than R6G. RhB is used as a fluorescent molecular thermometer,20-23 whereas R6G is not. The possibility of using shocked dye lifetimes to probe microstructured media, which has to be done on a single shot, was motivated by recent successes in fluorescence lifetime imaging microscopy (FLIM),1 and the difficulties we foresee in using redshift and intensity-loss for such measurements. To measure time- and position-dependent shock-induced redshifts, we would have to obtain a stream of wavelength-resolved emission spectra at every spatial location, which is impractical.

Intensity loss could be measured at many spatial locations using a fast

multielement detector such as a photodiode array, a photomultiplier array24 or a streak camera. However, shock intensity measurements are complicated by material moving around during the shock. The fast translational25 (several µm ns-1) and rotational26 motions of the microstructure, and shock-induced changes in emissivity, make it difficult to interpret intensity changes in a dynamically-evolving shocked medium. The singlet lifetime may be directly related to the chemical and physical environment, and it is insensitive to the intensity of the excitation pulses or changes in the geometry or emissivity that affects the detection efficiency,1 provided the pump pulses are kept at reasonable intensities below the thresholds for stimulated emission or singlet annihilation, as we did here. Since lifetime measurements need not be spectrally resolved (although that was done here), high-speed spatially-dependent variations of emission lifetimes could be captured using a fast detector array or a streak camera. With a streak camera, one could measure spatially dependent lifetimes along a line running through the microstructure. 2. Shocks in PMMA

5 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 29

When a flyer plate strikes a sample whose initial density is ρ0, the flyer/sample interface will begin to move at velocity Up, launching a shock at velocity Us.25 We measure Up on every shot with a photon Doppler velocimeter (PDV),27 and we can use Up to determine pressure P and estimate temperature T as follows. Us is related to Up by the Hugoniot,25 which for PMMA is approximately linear in the 0-10 GPa range,28,29 U s = mU p + b .

(1)

Based on our previous measurement of the Hugoniot for solvent-cast PMMA thin films, b = 2.6 km·s-1 and m = 1.2.29 The fractional change in volume ∆V is given by,25

∆V =

Us −U p Us

.

(2)

The pressure is,25

P=

∆Vb 2 V0 ( m∆V − 1)

2

,

(3)

where V0 = 1/ρ0 is the specific volume.25 Note that this method of determining P is based solely on conservation of momentum and relies on no approximations. When the initial temperature is T0, the temperature during the shock T1 is given by,25 T1 = T0 e

Γ(V ) ∆V

V1

+∫

V0

f (V ) e CV

Γ(V )∆V

dV ,

(4)

where Γ(V) is the Grüneisen parameter, Cv the constant-volume heat capacity and f(V) =

[ 12 (V0 − V1) dVdP + 12 P] .

The first term describes the temperature increase for a reversible

adiabatic compression and the second term adds a correction for the additional heating due to irreversible single-stage shock compression. Calculations using Eq. (4) are at best approximate 6 ACS Paragon Plus Environment

Page 7 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

because Γ(V) is not known to a high degree of certainty, and T is exponentially-sensitive to Γ(V). A common way of approximating Eq. (4) is to assume Γ(V)/V = constant, and Cv(T) varies little in the computed temperature range.25 While a temperature-independent Cv is appropriate for atomic solids, for PMMA Cv increases with temperature. However the increase is approximately linear in T,30

Cv (T ) = 4.40 J ⋅ Kg −1 ⋅ K −2 + 143J ⋅ Kg −1 ⋅ K −1 ,

(5)

so in Eq. (4) we used the average value of Cv in the temperature interval. Figure 2a shows computed results for PMMA using Γ = 0.6.31 With a 4 GPa shock, ∆T ≈ 100K, and with a 10 GPa shock, ∆T ≈ 300K. A shock is more than just pressure and temperature. There are dynamical effects as well, particularly in a high-molecular-weight polymer such as PMMA (MW = 1,000,000) which has a sluggish mechanical response strongly dependent on shock duration.12 Increasing shock duration at constant pressure produces higher polymer densities.12 Static compression in a diamond-anvil cell produces higher densities9 than any duration shock of equivalent pressure. To illustrate the sluggish PMMA response and the R6G shock photophysics, look at Figs. 2b,c which summarize results from ref. 11. Figure 2b shows the redshift of free R6G dye in PMMA using 75 µm thick Al flyers at 2 km·s-1 that produce 8.6 GPa shocks. The steadily-driven part of the shock lasted 18 ns, and the shock load was over by about 40 ns (c.f. Fig. 3c below). The redshift rose continuously during the 18 ns steadily-driven shock. After the steady shock drive ended, the redshift continued to increase, but more gradually, to a peak value of 40 nm at 40 ns. Afterwards, the redshift decayed with a time constant of ~70 ns. The redshift dynamics in Fig. 2b reflect the instantaneous R6G electronic response to time-dependent shock-induced changes in PMMA density. Figure 2c shows the shock-induced intensity loss in the same experiment, 7 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

where R6G was continuously pumped by a 250 ns laser pulse.

Page 8 of 29

The intensity-loss time

dependence was different from the redshift due to lag times produced by the dye’s photophysical response.11 The intensity loss rose a bit more slowly,5,11 and the decay was quite a bit slower than the redshift decay. When PMMA was deoxygenated, by heating it in a vacuum above the glass transition prior to the shock experiments, the intensity loss did not decay at all on the experimental time scale.11 These results were interpreted to indicate that the intensity loss was caused by shock-enhanced intersystem crossing, which trapped the R6G excitations in the dark long-lived triplet state.11 The intensity-loss decay reflects ground-state repopulation occurring at the triplet lifetime, and that lifetime increased when oxygen was removed. The significant amount of shock-enhanced intersystem crossing is interesting, because under ambient conditions the R6G quantum yield is so high32 that there is hardly any intersystem crossing. 3. Experimental The experimental apparatus is depicted schematically in Fig. 3. The idea is to push a button to fire a single shot of the flyer plate launch laser and the femtosecond Ti:sapphire probe laser, and trigger the streak camera once, while continually running the launch and probe lasers at 10 Hz and 1 kHz respectively, to maintain constant thermal loads on the laser rods. The flyer plate launch system with PDV has been described previously.27 The flyers were Al foils 75 µm thick and 500 µm in diameter, launched by 20 ns Nd:YAG laser pulses with uniform “tophat” beam profiles created by a diffractive beam homogenizer.10,27 The flyer speeds and arrival times were quite reproducible. Prior PDV studies,27 indicate the flyer speeds were reproducible to within ±0.6%, and the arrival time variation at the target was ±1 ns. The femtosecond laser was a modified Quantronix Integra-C 2.0. It had a mode-locked fiber laser oscillator (Calmar Laser, Mendocino 780) operating at 50 MHz. The 50 MHz signal was divided

8 ACS Paragon Plus Environment

Page 9 of 29

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

to 1 kHz to trigger the chirped-pulse laser amplifier, which output 130 fs pulses at 785 nm with 2.0 mJ energies. The 1 kHz output passed through an optical chopper (Thorlabs MC2000) with a custom chopper wheel (Fig. 3a), to reduce the repetition rate to 100 Hz, slow enough that a mechanical shutter (Vincent Associates) could be used to select individual pulses. A portion of each 785 nm pulse was frequency-doubled in a nonlinear crystal to generate 392 nm pulses to excite R6G. A photodiode observing the 100 Hz pulses triggered a Stanford Research Systems DG645 digital delay generator, which triggered the Nd:YAG launch laser at 10 Hz. A homebuilt pulse generator synchronized with the digital delay generator was then able, when a button was pushed, to open up mechanical shutters in the launch laser beam (NM laser) and in the femtosecond laser beam (Vincent Associates), and to enable a single trigger to the streak camera. The streak camera impulse response function, measured with a femtosecond pulse (c.f. Fig. 6a), had a full-width at half-maximum of 100 ps. Figure 3c shows a typical PDV result27 for the impact of a 75 µm thick flyer plate launched at 1.1 km·s-1. The PDV analysis here used the fringe-counting method described previously,27,33 which gave a time resolution of ~2 ns. Time t = 0 indicates the instant of flyer plate impact with PMMA, as determined by the PDV data, and due to the PDV time resolution, this instant was uncertain by about ±1 ns. The sudden decrease in flyer velocity at the instant of impact occurred within the apparatus time resolution. Thus variations in flyer flatness and tilt that would spread the velocity change over a finite time interval were quite small, no more than 2 µm given an ~1 km·s-1 velocity.

In Fig. 3c, after impact at t = 0, the flyer/sample interface moved at velocity Up = 0.9 (±0.02) km·s-1, giving a shock pressure P = 4 (±0.12) GPa. Up remained constant for 18 ns, which was the duration τsh of the steadily-driven shock in PMMA.27 After that time interval,

9 ACS Paragon Plus Environment

The Journal of Physical Chemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 29

inertia combined with polymer chain relaxations9,12,34 and edge rarefaction waves10 caused the PMMA surface to move with a steadily decreasing velocity for the next 20 ns. The 18 ns steadily-driven shock was followed by a decaying shock that ended within 40 ns of impact. The sample arrangement, which was almost the same as used previously,5,11,12 is shown in Fig. 3b. The sample on a 50 x 50 x 6.35 mm3 glass substrate was a 15 µm thick layer of dye/PMMA atop a 50 µm thick layer of poly vinyl alcohol (PVA). The PVA acted as a cushion that delayed and attenuated shock reflections from the glass substrate, and the one change we made in our sample design was to increase its thickness from 30 µm to 50 µm to make a better cushion. The PMMA layers contained either 5 mM R6G dye or 1 µm diameter spherical particles of SiO2-encapsulated R6G.

The impedance mismatch between silica and PMMA

means a shock passing through PMMA initially produces a lower pressure in the microparticles. The shock has to ring up in the silica particles before silica and PMMA are at the same pressures. The ring-up takes a few multiples of the shock transit time across the 1 µm silica particles, and the transit time is