libMCXray: A Monte Carlo Simulator for Signal Analysis inside Dragonfly Software


S. Rudinsky,1 M. Gendron,2 N. Piché,2 sam.rudinsky@steaminstruments.com


Abstract: Monte Carlo simulations are commonly used in elemental quan- tification using energy-dispersive spectroscopy (EDS). Here, the Monte Carlo program MC X-ray was incorporated into the image processing soft- ware Dragonfly by Object Research Systems (ORS) as a simulation library. The simulation program has been transformed into a complete micro- scope simulator where the tools of Dragonfly allow complex voxel-based geometries to be constructed, and the electron beam and detectors can be freely placed inside the 3D space. Computation times of simulations have been improved drastically through new data structures and parallel- ization. Simulations of backscattered electron imaging and EDS mapping are presented here to demonstrate the capabilities of this new library.


Keywords: scanning electron microscopy, energy-dispersive spec- troscopy, segmentation, Monte Carlo simulations, Dragonfly software


Introduction Monte Carlo simulations are widely used in applications


related to both electron beam imaging and microanalysis [1]. Tey provide information such as the interaction volume of incident electrons and X-ray distributions [2], and they are use- ful in cathodoluminescence emission measurements [3] and mass thickness estimations [4,5]. Tey also compute backscat- tered and secondary electron yields, which are important for image simulations used to estimate signal variations [6,7] and secondary election emissions of electrode materials [8]. Finally, a crucial use of Monte Carlo simulations is to perform accurate elemental quantification from energy-dispersive spectroscopy (EDS) data. Matrix correction factors can be computed using Monte Carlo simulations and then used to calculate accurate compositional information from X-ray intensities [9–11]. Tis allows quantification to be done without the need for standards or pre-measured experimental databases [10]. Te most widely used programs available are PENEPMA,


CASINO, and DTSA-II [12–14]. While these programs have been well adapted for common uses in electron microscopy and microanalysis, there exist some limitations to their usability. Te most notable of these are the computation time and restric- tions on sample and microscope geometries. For example, sim- ulations of electron numbers on the order of 105


at a single point


may take approximately a minute on a desktop computer [15]. Tese long computing times can make simulations of backscat- tered electron (BSE) images or EDS maps unfeasible in a rea- sonable amount of time. Te limitations in terms of geometry impact the usability of these programs on complex materials. Important obstacles arising from non-flat sample geometries can strongly impact accurate elemental quantification [16,17]. To simulate such materials, freedom in terms of sample and microscope geometry and placement must be possible. Here, we present the incorporation of the existing Monte Carlo soſtware MC X-ray [18] into the image processing soſtware


40 doi:10.1017/S1551929520001315


Dragonfly developed by Object Research Systems (ORS) Inc. [19]. Te class system of Dragonfly allows substantial freedom in terms of sample geometry and placement. Furthermore, additions have been made to turn MC X-ray into a full micro- scope simulation tool capable of reproducing some elements of an electron microscope. Multiprocessing has also been imple- mented to decrease computing times substantially, allowing more complex problems to be easily evaluated. Te manuscript is structured as follows. First, the details and usability of the program are described with reference to previous Monte Carlo soſtware. Ten applications of the library to BSE image simula- tions and elemental quantification are presented. Finally, some concluding remarks are made about the method.


Materials and Methods MC X-ray is a Monte Carlo-based simulation package [18]


that was devised as an extension of CASINO [13,20] and Win X-ray [21], specific for X-ray microanalysis. Scattering events are modeled based on a stochastic process where electrons are simu- lated using a forward scattering random walk. An electron is ini- tiated, and uniform random numbers are generated and used in cross-sectional models to determine the particle’s path through the sample material. Te physical models, methodology, and cal- culations used are described in other work [21]. Te previously stand-alone MC X-ray program was incorporated into the image processing soſtware Dragonfly as a feature plug-in. Te base code was written in C++, and the interface used to gather the input and call the simulation functions was written in Python 3. Sample generation is done using the Multi-region of inter-


est (ROI) class inside Dragonfly. A phantom material is created as a MultiROI where each subgeometry is its own ROI. Te structures are voxel-based where the dimensions of the voxels are chosen by the user. Once a phantom has been defined, labels may be assigned to each ROI. Each label contains a property where, for the purpose of libMCXray, these properties represent the constituent elements of each region. Te labels are vectors whose lengths are the number of regions and whose values are the composition of the elements in the indexed region. Tus, a material may be generated with a multitude of elements in a variety of regions with varying compositions. MultiROIs can be created by the user or generated from other image data such as stacks of images, which can be compiled into a 3D structure. Te simulation environment comprises an electron beam,


a sample material, and a series of detectors. Te electron beam is above the sample material normal to its surface. Te user input parameters of the beam are the incident electron energy, incident current, and working distance. Te rasterizing across the sample surface is done by displacing the beam across the


www.microscopy-today.com • 2020 September M. Marsh,2 and R. Gauvin3 1Steam instruments, 931 E. Main St., Ste. 3, Madison, WI, 53703-2955


2Object Research Systems, 101-760 Rue Saint-Paul Ouest, Montreal, QC, H3C 1M4 3McGill University, 3610 University, Montreal, QC, H3A 0C5


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38  |  Page 39  |  Page 40  |  Page 41  |  Page 42  |  Page 43  |  Page 44  |  Page 45  |  Page 46  |  Page 47  |  Page 48  |  Page 49  |  Page 50  |  Page 51  |  Page 52  |  Page 53  |  Page 54  |  Page 55  |  Page 56  |  Page 57  |  Page 58  |  Page 59  |  Page 60  |  Page 61  |  Page 62  |  Page 63  |  Page 64  |  Page 65  |  Page 66  |  Page 67  |  Page 68  |  Page 69  |  Page 70  |  Page 71  |  Page 72  |  Page 73  |  Page 74  |  Page 75  |  Page 76  |  Page 77  |  Page 78  |  Page 79  |  Page 80