Electronic Supplementary Information Development of an advanced multiwavelength emission detector for the analytical ultracentrifuge

An advanced design of the analytical ultracentrifuge with multiwavelength emission detection (MWE-AUC) is presented which offers outstanding performance concerning the spectral resolution and range flexibility as well as the quality of the data acquired. The excitation by a 520 nm laser is complemented with a 405 nm laser. An external spectrograph with three switchable tunable gratings permits optimisation of the spectral resolution in an order of magnitude range while keeping the spectral region broad. The new system design leads also to a significant reduction of systematic signal noise and allows the assessment and control of inner filter effects. Details regarding the very large signal dynamic range are presented, an important aspect when studying samples in a broad concentration range of up to five orders of magnitude. Our system is validated by complementary studies on two biological systems, fluorescent BSA and GFP, using the commercial Optima AUC with absorbance detection for comparison. Finally, we demonstrate the capabilities of our second generation MWE-AUC with respect to multiwavelength characterisation of gold nanoclusters, which exhibit specific fluorescence depending on their structure. Overall, this work depicts an important stepping stone for the concept of multiwavelength emission detection in AUC. The MWE-AUC developed, being to our knowledge the first and sole one of its kind, has reached the development level suitable for the future in-depth studies of size-, shape- and composition-dependent emission properties of colloids.


ADC & tuneable
PMT gain 1 Required volume for sedimentation velocity experiments to provide 85-90% cell filling with well visible meniscus.For sedimentation equilibrium using special centrepieces, required sample volume can be up to 4 times smaller.1a Estimated reduction of the signal amplitude of a 1.5 mm compared to 3 mm centrepiece is ~ 20%.
2 External laser can be easily exchanged by the operator.There are no limitations on the laser dimensions, power consumption, cooling, etc. 3 Variable optical attenuator allows tuneable power reduction by two orders of magnitude without changing other parameters of the laser. 4External spectrograph, having no limitation on its size, provides a unique flexibility of spectral range and spectral resolution by using different gratings (three on a turret with easy exchange of turrets by the operator).The central wavelength is quickly tuneable also during measurements by changing the grating angle using computer control, as well as the range by changing the grating used.In addition to the EMCCD camera, a second, e.g., NIR or UV-optimized camera, photomultiplier or any other detector can be installed on the second spectrograph output with computer control of the output used. 5Resolution can be improved by postprocessing of the 2D camera image, as well as by reducing the tuneable entrance slit with corresponding reduction of the signal amplitude. 6Without electron multiplication (EM).The dynamic range can be further increased by integration over several camera pixels.EM gain up to x1000 allows single photon counting.

Figure S1 .
Figure S1.Block scheme of the triggering.

Figure S4 .
Figure S4.Absorbance spectra for different concentrations of Coumarin 153 measured prior to AUC measurements with a Varian Cary 100 spectrophotometer (2.0 nm slit width, 10 mm optical path, 1 nm spectral resolution).Dashed line marks the 405 nm excitation laser line.

Figure S5 .
Figure S5.Radial scan of 0.1 mg/L Coumarin 153 dissolved in ethanol showing the meniscus for various wavelengths.Data are recorded with 10 µm steps at 5,000 rpm.The 405 nm laser was used for excitation, which provides a focal spot smaller than the 520 nm laser.a) Radial resolution of 30 µm as the distance of the edge rise from the 10 % to the 90 % level.b) Distortions of the sedimentation profile in the meniscus region.

Figure S6 .
Figure S6.Signal amplitude of Coumarin 153 samples diluted in ethanol to target concentrations of 0.001 mg/L and 0.0001 mg/L measured at different exposure times to illustrate the linearity of the Gen2 MWE-AUC.Other than in Figure 4a) in the main manuscript, the linear y-axis allows a proper visualisation of error bars for very low signal amplitudes.

Figure S7 .
Figure S7.a) Normalised absorption and emission spectra of GFP diluted in a 20 v% glycerol/PBS buffer.Absorption and emission are shown for GFP concentrations of 5 µM and 2.5 µM, respectively.b) Normalised absorption and emission spectra for 1 mg/L Coumarin 153 dissolved in ethanol.Absorption spectra were measured with Varian Cary 100 spectrophotometer (2.0 nm slit width, 10 mm optical path, 1 nm spectral resolution).Dashed lines mark the 405 nm excitation laser line.

Figure
Figure S8.a) Emission spectrum of 10 mg/L Coumarin 153 dissolved in ethanol recorded at z-position 12,500 µm.Straight lines indicate the selected wavelengths plotted in b) and c).b) Z-scans for the wavelengths shown in a) demonstrating distortions by the inner filter effects.c) Normalised z-scans of Coumarin 153 at several wavelengths, demonstrating the wavelength independence of the Gen2 MWE-AUC.The asymmetry visualises the apparent inner filter effects.Dashed lines show the z-scans for 455 nm and 470 nm demonstrating additionally the weak but still apparent secondary inner filter effect.

Figure S9 .
Figure S9.Fit results and residuals for GFP(a, b)  andF-BSA (c, d)  measured with the Optima AUC with the absorption detection system (a, c) and Gen2MWE-AUC (b, d).The first 350 scans were considered for evaluation.The 520 nm laser was used for excitation in the MWE-AUC measurements.GFP was diluted in a glycerol/PBS buffer and was measured at 5 µM in the Optima AUC and 2.5 µM in the MWE-AUC (data shown are for emission at 570 nm).The comparably slow sedimentation of GFP can be explained by the presence of glycerol in the sample, whose density and viscosity are increased compared to water.F-BSA was diluted in a Tris (12 mM)/NaCl (15 mM) buffer and was measured at 0.45 µM in both devices.F-BSA data by the MWE-AUC were evaluated at 550 nm.Note: Noise spikes observable in d) are due to "cosmic rays" being recorded by the EMCCD camera.Such environmental pollution only occurs sporadically and can be corrected for during data processing.Here the data were not post-treated and are plotted as recorded.All protein measurements were carried out at 40,000 rpm and 20°C.

Figure S10 .
Figure S10.2D analysis of sedimentation velocity experiments ofAuNCs performed at 40,000 rpm, 20°C  and an excitation at a) 405 nm and b) 520 nm.Data were integrated over 1.5 nm spectral bandwidth.For a) a partial specific volume of 0.65 cm 3 /g and frictional ratio of 1.4 was used for fitting the data instead of the best-fit values (0.547 cm 3 /g and 1.0, respectively) used for deriving the distributions shown in Figure9ain the main manuscript.Both values were chosen manually in such way that the wavelength shift for each species is as small as possible.For b) the best-fit values for the partial specific volume (0.63 cm 3 /g) and frictional ratio (1.0) were used as no pronounced wavelength shift was observed, which is identical to the results reported in Figure9bin the main manuscript.Both figures shown here display the nonregularised distributions.

Figure S11 .
Figure S11.Fit results for sedimentation profiles and residuals of global analysis of AuNCs in SEDPHAT at wavelengths of a) 700 nm with bandwidth 5 nm, b) 760 nm with bandwidth 10 nm, and c) 825 nm with bandwidth 5 nm.The excitation was at 405 nm.Sedimentation coefficients of 3.35 S, 3.97 S and 4.82 S were obtained.

Table T2 .
Specifications of the Gen2 MWE-AUC and comparison with the AU-FDS.