Context. The behaviour of molecules in space is to a large extent governed by where they freeze out or sublimate. The molecular binding energy is therefore an important parameter for many astrochemical studies. This parameter is usually determined with time-consuming experiments, computationally expensive quantum chemical calculations, or the inexpensive yet relatively inaccurate linear addition method. Aims. In this work, we propose a new method for predicting binding energies (BEs) based on machine learning that is accurate, yet computationally inexpensive. Methods. We created a machine-learning (ML) model based on Gaussian process regression (GPR) and trained it on a database of BEs of molecules collected from laboratory experiments presented in the literature. The molecules in the database are categorised by their features, such as mono- or multilayer coverage, binding surface, functional groups, valence electrons, and H-bond acceptors and donors. Results. We assessed the performance of the model with five-fold and leave-one-molecule-out cross validation. Predictions are generally accurate, with differences between predicted binding energies and values from the literature of less than ±20%. We used the validated model to predict the binding energies of 21 molecules that were recently detected in the interstellar medium, but for which binding energy values are unknown. We used a simplified model to visualise where the snow lines of these molecules would be located in a protoplanetary disk. Conclusions. This work demonstrates that ML can be employed to accurately and rapidly predict BEs of molecules. Machine learning complements current laboratory experiments and quantum chemical computational studies. The predicted BEs will find use in the modelling of astrochemical and planet-forming environments.

Despite the numerous existing (semi)automated workflows for image segmentation of electron microscopy pictures of nanoparticles for statistical size and shape determination the prevalent approach to particle counting still is doing so in cumbersome manual fashion. Here, we present an easily implementable, low entry barrier workflow for nanoparticle segmentation, which eliminates the need for manual particle counting. It is based on the recently released segment anything model and widely distributed, well maintained, python libraries. We explore the impressive zero shot performance of the segment anything model and present approaches for subsequent filtering of outputs to minimize over and under segmentation on a range of different electron microscopy images of nanoparticles. Furthermore, we introduce a novel methodology for handling partial overlap between nanoparticles, which comprise one of the biggest obstacles for many automated segmentation algorithms. Our presented workflow is easily adaptable, and we encourage the community to further build on the work we present here.

Despite the numerous existing (semi)automated workflows for image segmentation of electron microscopy pictures of nanoparticles for statistical size and shape determination the prevalent approach to particle counting still is doing so in cumbersome manual fashion. Here, we present an easily implementable, low entry barrier workflow for nanoparticle segmentation, which eliminates the need for manual particle counting. It is based on the recently released segment anything model and widely distributed, well maintained, python libraries. We explore the impressive zero shot performance of the segment anything model and present approaches for subsequent filtering of outputs to minimize over and under segmentation on a range of different electron microscopy images of nanoparticles. Furthermore, we introduce a novel methodology for handling partial overlap between nanoparticles, which comprise one of the biggest obstacles for many automated segmentation algorithms. Our presented workflow is easily adaptable, and we encourage the community to further build on the work we present here.