Intense mechanoluminescence and gas phase reactions from the sonication of an organic slurry.

Mechanoluminescence (ML, also known as triboluminescence or fractoluminescence) is the emission of light due to the application of mechanical stress to a crystal.1 This light is the result of a dielectric breakdown of the intervening gas and the crystal caused by internal electrification at the fracture plane of a piezoelectric crystal. When ML is induced by such mechanical means as grinding, scratching, or cleaving, a spectrum is observed consisting of line emission of the gas present and sometimes the crystal itself.1,2 In fact, the first report of ML goes back to 1605, when Francis Bacon noted that lumps of sugar emitted light when scraped.3 When a liquid is irradiated with high-intensity ultrasound, micron-sized bubbles can be driven into violent collapse, sending shockwaves outward into the liquid.4 If small solid particles are suspended in the liquid, the shockwaves can accelerate them to high velocities (several hundred m/s vs cm/s for manual grinding) causing intense interparticle collisions. We report here the initiation of ML using acoustic cavitation and observe extensive atomic and molecular emission not previously seen during a mechanoluminescent event. In addition, we have the first evidence of chemical reactions occurring during a ML event. We recently reported ML from sonication of slurries of piezoelectric organic crystals in long chain alkanes,5 and the N2 ML emission from sonication of sucrose was an order of magnitude more intense than that from grinding. We report here that sonication of a slurry of resorcinol in dodecane sparged with N2, the N2 (CΠu BΠg) and the N2 (BΣu XΣg) emission spectrum (Figure 1), is more than 1000-fold more intense than that from manual grinding. We have also observed new emission lines not previously seen during ML (including CO, CO+ and CO2, H, C2, and CH) showing the existence of chemical reactions during the ML event. When a slurry of resorcinol is sonicated in the presence of He, several new atomic and molecular emission series are observed, in addition to the crystal luminescence and strong He discharge. These include He+ (∼77 eV above the 1s state of He), multiple transitions from the Swan band of C2 (dΠg aΠu), the CH (A2∆ X2Π) transition, and the H R line (Figure 2A). When the same slurry is sparged with O2, the same crystal luminescence, C2 and CH discharges are still present, and in addition, emissions from O, CO (b3Σ a3Π), (C1Σ A1Π) and (B1Σ A1Π), CO+ (B2Π X2Σ), and CO2 (A2Π X2Π) are newly observed (Figure 2B). The emission of C2, CH, CO, CO+, and CO2 has never been reported before from any ML event. It should be noted that the Swan band of C2 as well as the CH transition at 431 nm are commonly observed in simple sonoluminescence (SL) of hydrocarbons,6 but the lines from such SL are substantially broader and Very much weaker than the observed ML (cf. Supporting Information, Figure S1). The multitude of new features arising in the ML spectra could have arisen from either the liquid or the solid. The results of sonication of slurries of resorcinol in dodecane (C12H26) versus perdeutero-dodecane (C12D26) are shown in Figure 3. In the dodecane slurry, only CH emission is observed, whereas in the perdeutero-dodecane slurry, only CD emission is present. This indicates that the carbon-containing small molecules in the ML spectra are from the liquid and not the crystal. The low vapor pressure of the dodecane and hexadecane, however, presents us with a paradox because it seems unlikely that the new emitting species can arise directly from the vapor of the liquid itself. Sonication of long chain alkanes, however, is known to produce small hydrocarbons (i.e., CH4 and C2H2) and H2; moreover, when a reactive gas is present (such as N2 or O2), other sonolysis products are formed, such as CN and CO2. We hypothesize that the observed CH, C2, and H originate from the dielectric discharge of sonolysis products and not directly from the low volatility liquid hydrocarbons themselves. To test this, a slurry of resorcinol and dodecane was sparged with He and then sonicated in a sealed cell without further sparging; the intensity of the CH and H emission increased upon multiple sonications, which is consistent with our hypothesis that the volatile sonolysis products of the liquid play a major role in the discharge emission. The origin of the CO, CO+, and CO2 and their emission must arise during the dielectric breakdown from chemical reactions, either of CO2 produced during sonolysis or of O2 with small hydrocarbon sonolysis products (e.g., CH4 or C2H2). When a slurry of resorcinol in hexadecane sparged with CO2 was sonicated, the CO2 (A2Π X2Π) and (A2Σ+ X2Π) emission bands were observed, with no visible emission from CO, CO+, or O, which are the main components of the ML from O2 sparged slurries containing resorcinol (Figure 2b). The production of CO and CO+ is therefore due to discharge reactions of small hydrocarbons with O2 and not Figure 1. ML from sonication of resorcinol crystals in hexadecane (10% w/w), sparged with N2 at 300 K irradiated with ultrasound at 20 kHz, 6 W/cm2. The spectrum contains N2 (CΠu BΠg) and N2 (BΣu XΣg) progressions (from 300 to 500 nm). The inset (∼1000-fold intensity enlargement) shows very weak emission from N2 (BΠg AΣu), which has not been reported for any prior mechanoluminescent system (presumably due to its inherently low emissivity). Published on Web 05/05/2007