Distribution of iron in size resolved aerosols generated by femtosecond laser ablation: Influence of cell geometry and implications for in situ isotopic measurements using LA-MC-ICP-MS

Laser Ablation (LA) Inductively Coupled Plasma Mass Spectrometry (ICP-MS) is a common and reliable method for the in situ chemical analysis in geosciences. In contrast, use of LA for analyzing naturally occurring mass dependent isotope fractionation in geological samples is not widely used because of the difficulties of differentiating laser induced isotope fractionation from naturally occurring mass dependent isotope fractionation. A critical aspect for accurate in situ stable isotope analysis is the chemical and isotopic composition, size, and morphology of aerosols generated by LA. We report on the iron mass distribution over the particle size distribution (PSD) of aerosols produced by femtosecond LA of magnetite and pyrite. A Photon Machines "Analyte" fs-G2 was used that provides τ~150fs pulses in the near UV (λ=263nm) with an adjustable repetition rate 1<f<250Hz. Ablations were performed under Helium (~0.5L.min^-1 outgoing flow) and two different cells were used: one cylindrical with a ~10s wash out time, the other (Photon Machines HelEx), with a wash out of less than 1s. For all experiments, aerosols were collected on Teflon filters using a MOUDI cascade impactor, according to their aerodynamic diameter, within a range of d_a<0.056μm to d_a>18μm (Marple, Rubow et al., 1991). Filters loads were dissolved in HCl (magnetite) or HNO₃ (pyrite) and iron concentration was determined spectroscopically using the ferrozine method or by isotope dilution mass spectrometery. The PSD for a given cell is similar for both pyrite and magnetite and is independent of fluence (1-3J.cm^-2). However, significant differences appear from one cell to the other. The cylindrical cell gives a unimodal distribution with a peak centered on d_a=0.18μm and spread from d_a=0.056μm to 0.56μm (83% of the total Fe mass). Using the Helex cell the PSD is bi modal with ~1/3 of the particles having a d_a<0.056μm in addition to the peak at d_a=0.18μm. Importantly we note that for a given mineral the Fe removal rate normalized to fluence was relatively constant (~2×10^-5μg Fe/shot for magnetite). The differences in aerosol residence times and transport flow patterns may produce the different PSD observed between the two cells (Koch, Walle et al., 2008). For example, the long residence and regular vortex patterns during transport may increase the probability of collision of particles and favor agglomeration of very small particles in the cylindrical cell. In contrast, in the Helex cell, the faster extraction along with turbulent flows may minimize the production of larger agglomerates. Ongoing studies will evaluate the Fe isotope composition and morphology of the aerodynamically size sorted particles in these different cells. Koch, J., M. Walle, et al. (2008). "Analysis of laser-produced aerosols by inductively coupled plasma mass spectrometry: Transport phenomena and elemental fractionation." Analytical Chemistry 80(4): 915-921. Marple, V. A., K. L. Rubow, et al. (1991). "A Microorifice Uniform Deposit Impactor (MOUDI): Description, Calibration, and Use." Aerosol Science and Technology 14(4): 434-446.