Logo of the Physikalisch-Technische Bundesanstalt

Electron emission spectra from gold nanoparticles – DESY PETRA III beamtime


Gold nanoparticles (AuNPs) are used both in radiation therapy and in diagnostics to increase the effectiveness of beams. This enhancement is mainly due to the secondary electrons which are emitted by AuNPs when they are traversed by ionizing radiation. Within the scope of a recent comparison of the simulation results of different Monte‑Carlo codes, discrepancies of several orders of magnitude were found in fluence spectra for X‑ray‑induced, low‑energy electron emissions of AuNPs [1].

For this reason, electron emission spectra from gold nanoparticles were measured over a lare electron energy range in a beamtime at DESY’s PETRA III beamline 22. These experiments aimed to obtain an initial dataset for benchmarking Monte‑Carlo simulations.

Four different samples were examined during this beamtime: two samples of gold nanoparticles with average AuNP diameters of 5.3 nm and 12 nm, respectively, and two reference samples consisting of a gold and carbon foil. The gold nanoparticles were deposited by sputtering and drop‑casting them onto a 50 nm thick carbon foil which covered an aluminum support. The aluminum support, in turn, was mounted on a standard DESY wedge‑shaped copper sample holder (Fig. 1).

Fig. 1: Distribution of the AuNPs deposited by drop‑casting them onto an aluminum sample holder coated with a carbon foil. The sample holder was mounted on a DESY wedge‑shaped standard copper target holder.

The electron emission spectra from the AuNPs and reference samples were recorded with the HAXPES spectrometer for photon energies above and below the gold L‑edges at 14.4 keV, 14.3 keV, 13.8 keV, 13.7 keV, 12 keV, and 11.9 keV. The measured energy range of the secondary electrons was between 100 eV and 9.5 keV. The angle‑resolved measurements were performed for photon incidence angles of 15° and 60° to allow the effective layer thickness of the AuNPs to be determined. Measurements were repeated at least two times to check reproducibility. To enable the results to be put on an absolute scale, additional measurements were performed to determine the energy dependence of the spectrometer transmission. Moreover, the photon flux and the beam profile were determined with a calibrated photodiode by scanning the beam across the edges of the diode.

Figure 2 shows an example of the electron emission spectra for the two AuNP samples measured. The measurement was done above the gold L1‑edge of gold at 14.4 keV and at an incidence angle of the photon beam of 60°. The spectra were evaluated with regard to the position, shape, width, and intensity of the lines. For example, the most intense peak at 2.5 keV is the Au‑L3 photopeak, and the sharp peak at 2.11 keV in the 5.3 nm spectrum is the Au‑M4N6N7 Auger peak. The broad structure between 4 and 5 keV in the 12 nm data is the so‑called Tougaard background arising due to inelastic scattering of copper target K‑shell photoelectrons from the sample holder [2,3]. Additional data analyses are in progress to establish to which extent the difference in the spectra between the two AuNP samples can be attributed to the different AuNP sizes.

Fig. 2: Comparison of the measured electron emission spectra for two different sizes of AuNPs at 14.4 keV above the gold L1-edge at a 60° incidence ngle.

The measurements have provided a preliminary dataset for benchmarking Monte‑Carlo simulations. In a second beamtime, it is planned to carry out further measurements with AuNPs on a thicker carbon substrate. This should allow the effect of the Tougaard background to be minimized and thus better‑quality data to be obtained. In the future, measurements may be carried out with a considerably higher resolution to investigate the contribution of plasmon excitation to the spectrum of the secondary electrons.


[01]    W. B. Li et al., “Physica Medica Intercomparison of dose enhancement ratio and secondary electron spectra for gold nanoparticles irradiated by X‑rays calculated using multiple Monte Carlo simulation codes”, vol. 69, no. December 2019, pp. 147–163, 2020.

[02]    R. Hesse and R. Denecke, “Improved Tougaard background calculation by introduction of fittable parameters for the inelastic electron scattering cross‑section in the peak fit of photoelectron spectra with UNIFIT 2011”, no. October 2010, pp. 1514–1526, 2011.

[03]    NIST X-ray Photoelectron Spectroscopy Database, NIST Standard Reference Database Number 20, National Institute of Standards and Technology, Gaithersburg MD, 20899 (2000), (retrieved 29.10.2020).


Opens local program for sending emailP. Hepperle, Department 6.3, Working Group 6.36