Challenges in measuring nanoparticles and microparticles by single particle ICP-QMS and ICP-TOFMS: size-dependent transport efficiency and limited linear dynamic range

Abstract

While spICP-MS has been used mainly to measure nanoparticles, it can also be used to measure microparticles. The transport efficiency of nanoparticles is typically independent of their size. However, the transport efficiency of microparticles can be particle size (mass) dependent as well as being dependent on the sample uptake rate and sample introduction system used. To measure both nanoparticles and microparticles a very large linear dynamic range (where signal intensity is linearly proportional to the measured analyte(s) mass within a very short measurement time (∼300 to 500 µs, the width of signals produced by an individual particle)) is needed. Deviations from linearity could occur due to incomplete particle vaporization or from signals that are beyond the instrument's ion detection system linear dynamic range. To characterize and determine the cause of nonlinearity we measured sets of nearly monodisperse engineered SiO₂ particles with diameters from 500 to 5000 nm and Au particles with diameters from 60 to 1500 nm. We found that by reducing the sensitivity (up to a factor of 269×) the upper end of the linear dynamic range, in particle size that produced signal intensities that were linearly proportional to the particle (analyte) mass, could be greatly extended. Not surprisingly, reducing the sensitivity increased the minimum size detectable particle. The results are consistent with SiO₂ particles as large as 5000 nm being completely vaporized in the ICP.

Fig. 1. Transport efficiency of engineered Au (), and SiO₂ () NPs and µPs as a function of the mass (fg) of each particle measured by spICP-QMS at using an uptake rate of 27 µL min⁻¹ (filled symbols) and 60 µL min⁻¹ (open symbols). Nominal particle size (nm) is shown next to each point. Error bars indicate ± one standard deviation of three measurements of transport efficiency.

Fig. 2. ²⁸Si⁺ signal in counts (left y-axis) and estimated counts per s (right y-axis) as a function of Si mass in particle produced by engineered SiO₂ particles measured by spICP-QMS using optimized sensitivity for particles with: (a) masses less than 800 fg and (b) masses less than 17 000 fg. The nominal particle diameter is shown next to each point. Filled symbols indicate particles used to calculate the linear regression line while open symbols indicate particles not used to calculate the linear regression line.

Fig. 3. ²⁸Si⁺ signal in counts (left y-axis) and estimated counts per s (right y-axis) produced by engineered SiO₂ particle suspensions measured by: (a and b) TOFWERK icpTOF-R and (c and d) Nu Instruments Vitesse spICP-TOFMS using optimized sensitivity for particles with: (a and c) masses less than 800 fg and (b and d) masses less than 17 000 fg. The nominal particle diameter is shown next to each point. Filled symbols indicate particles used to calculate the linear regression line while open symbols indicate particles not used to calculate the linear regression line.

Fig. 4. ²⁸Si⁺ signal in counts (left y-axis) and estimated counts per s (right y-axis) produced by engineered SiO₂ particle suspensions measured by a PerkinElmer NexION 350D spICP-QMS with highly reduced sensitivity (269× compared to optimized sensitivity). Nominal particle diameter is shown next to each point. Filled symbols indicate particles used to calculate the linear regression line while open symbols indicate particles not used to calculate the linear regression line.

Fig. 5. Particle number concentration (particles per mL) versus Si⁺ intensity of (a) 895 nm, (b) 1050 nm, (c) 2060 nm, and (d) 3170 nm SiO₂ particles measured by spICP-QMS with reduced sensitivity (269× compared to optimized sensitivity). The expected average Si⁺ intensity produced by 895 nm and 1050 nm particles () (based on their size and the sensitivity calculated from the 3170 nm average intensity) is shown on plots (a and b). The measured average Si⁺ intensity () is shown on plots (c and d). The signal threshold to detect a particle event () is shown on each plot. n_reduced is the number of particles detected using 256× reduced sensitivity. n_optim is the number of particles detected using the optimized sensitivity.

Fig. 6. ²⁸Si⁺ signal multiplied by the sensitivity reduction factor in counts (left y-axis) and estimated counts per s if no sensitivity reduction (counts per s multiplied by the sensitivity reduction factor) (right y-axis) produced by engineered SiO₂ particle suspensions measured by a PerkinElmer NexION 350D spICP-QMS with optimized sensitivity (), 13× reduced sensitivity () and 269× reduced sensitivity () for: (a) particles with masses less than 1000 fg, and (b) particles with Si masses less than 17,00 fg. A linear regression line () was calculated using the Si⁺ intensity x sensitivity reduction factor and mass of all plotted particles.

Fig. 7. ²⁸Si⁺ signal in counts produced by engineered SiO₂ particles measured by spICP-QMS using 182× reduced sensitivity. The nominal particle diameter is shown next to each point. Filled symbols indicate particles used to calculate the linear regression line while open symbols indicate particles not used to calculate the linear regression line.

Fig. 8. ²⁸Si⁺ signal in counts (left y-axis) and estimated counts per s (right y-axis) produced by SiO₂ particle suspensions measured using: (a and b) TOFWERK icpTOF-R and (c and d) Nu Instruments Vitesse spICP-TOFMS using reduced sensitivity for particles with diameters: (a and c) up to 2060 nm and (b and d) up to 3170 nm. icpTOF-R sensitivity was reduced 11× and Vitesse sensitivity was reduced 30× from the optimized sensitivity on each instrument. Nominal particle diameter is shown next to each point. Filled symbols indicate particles used to calculate the linear regression line while open symbols indicate particles not used to calculate the linear regression line.