I am interested in protoplanetary discs, their evolution and structure, both from a theoretical and observational point of view. I combine modeling of discs using state-of-the-art tools (such as MCMax, Torus and DALI) with (sub-)mm ALMA observations.
Publications- Boneberg, D.M.; Facchini, S; Clarke, C.J.; Ilee, J.D.; Booth, R.A.: An extremely truncated circumstellar disc: a precursor to TRAPPIST-1? (subm.) - Haworth, T.J.; Ilee, J.D.; Forgan, D.H.; Facchini, S.; Price, D.J., Boneberg, D.M., et al.: Grand challenges in protoplanetary disc modelling (major review article) - Boneberg, D.M.; Panić, O.; Haworth, T.J.; Clarke, C.J., Min, M.: The midplane conditions of protoplanetary discs: a case study of HD163296 - Boneberg, D.M.; Dale, J.E.; Girichidis, P.; Ercolano, B.: Turbulence in giant molecular clouds: the effect of photoionization feedback They can be found here.
Spectral energy distribution of a star with circumstellar disc. Figure credit: Dullemond et al. 2016.
Protoplanetary discs consist of gas (H2, CO etc) and dust (e.g. silicates, carbonaceous species,..). Observations of the Interstellar Medium suggest that the gas-to-dust mass ratio there is g/d~100, therefore this value is often also adapted in protoplanetary discs. The gas governs the dynamics of the dust, whereas the dust sets the disc temperature structure by providing the opacity to intercept stellar light and re-radiate it. As H2 (which is the most abundant component of the gas in discs) cannot be directly observed due to its lack of a dipole moment, other molecules (such as CO) are usually employed instead. The mass of gas can then be derived by taking into account an abundance of CO with respect to H2.
Protoplanetary discs are the birthplaces of planets, and it is therefore important to study their properties, especially the conditions in the disc midplane regions where planets are understood to form. In order to study the density and temperature structures of protoplanetary discs, a combination of high-resolution observations and in-depth modelling is necessary. In my studies, I use e.g. observations from the ALMA (Atacama Large Millimeter/submillimeter Array) observatory in Chile, providing data of unprecedented resolution both spectrally and spatially. These observations are then combined with state-of-the-art computer models of discs. In order to model discs, I have been using a combination of MCMax and TORUS as well as DALI and a dust evolution code developed by Dr Richard Booth.
Protoplanetary discs around stars leave a typical signature in the Spectral Energy Distribution (SED). In order to compose an SED, the flux coming from an object is measured in different wavelength and then plotted as a function thereof. For an isolated star, the SED is a simple black body, however the presence of a disc around the star leads to an excess in infrared emission. This can be seen in the schematic depiction of the SED. The SED is a tracer of the dust content of a protoplanetary disc.
In general, SEDs of discs crucially depend on the disc properties: For optically thin emission, the dust mass of the disc influence how high the fluxes are (more dust leads to higher fluxes). The gas on the other hand determines the motion and location of the dust grains in the disc, lofting grains to disc regions higher up if the gas is very turbulent. Another parameter influencing the SED are the minimum and maximum dust grains sizes, as well as their composition. In order to break the degeneracy between the effect of all these parameters, a combination of modelling steps is necessary.
A truncated disc around a very low mass star?
Schematic depiction of the two plausible scenarios: Radial drift and truncation.
In particular, my collaborators and I have studied the SED of a disc around a very low mass star with approximately 0.1Msun. From the Rayleigh-Jeans tail of the SED, we derive that the dust disc is very truncated, on the order of 1AU (which is the distance that the Earth has from the sun!). We propose two different scenarios to explain this feature (see the schematic picture):
1) Both the dust and the gas components of this disc are truncated, for example due to a dynamical interaction with another star. 2) Only the dust component appears to be truncated: this could be caused by radial drift of dust grains towards the star.
We propose that both of these scenarios are possibly and match the observational criteria given by the SED. In order to distinguish further between these two scenarios, we propose in-depth ALMA observations to pin down the radial extent of gas and dust. We have submitted a paper on this project (Boneberg et al. 2017).
The midplane conditions of protoplanetary discs
As the midplane regions of protoplanetary discs are important for the formation of planets, it is important to understand the temperature and density structures there in detail. One tracer of the disc midplane is a carbon-monoxide isotopologue C18O, whose emission is mostly optically thin throughout discs. CO (and its isotopologues) leave the gas phase and freeze out onto dust grains at temperatures of about 20K. The radial location outside of which this happens is called the 'snowline' and can be used to probe the temperatures in discs (see figure).
In this project, we have combined modelling of the C18O and dust emission from the protoplanetary disc HD163296 with ALMA observations thereof. We find that the models have to match three criteria: 1) The model SED has to match the observed SED at all wavelengths. This allows to get information on the dust component of the disc. 2) The models have to comply with the observed snowline, i.e. they have to have midplane temperatures of about 20K at a disc radius of ~90AU. 3) The C18O emission of the models has to match the one observed with ALMA.
From our modelling, we find that our models favour a rather low gas-to-dust ratio of ~20 when we assume a CO abundance of ~10-4 (as is generally assumed for the ISM and discs). This low g/d ratio could hint at the fact that the CO abundance might be changed by processes in discs, which has to be taken into account when deriving gas masses from the CO emission. Overall, we find that the combination of all three diagnostics (SED, snowline and C18O emission) is important in order to overcome degeneracies in disc modelling. This is described in detail in our paper Boneberg et al. 2016.
Turbulence in Giant Molecular Clouds
Before working on protoplanetary discs, I was studying objects on larger scales than protoplanetary discs: Giant Molecular Clouds (GMCs). These are regions within a galaxy where star formation takes place. GMCs are observed to be turbulent, but in order to prevent the decay of turbulence, some kind of driving mechanism is needed. There are many candidates for driving turbulence in GMCs. On large scales, these include amongst others the injection of energy by external accretion flows, turbulence driven by gravity or density waves in spiral arms or cloud-cloud collisions. Within the clouds themselves, photoionizing radiation (expanding HII regions), supernovae feedback, stellar winds/outflows and radiation pressure are discussed as drivers for turbulence.
In the project, we have studied the effect of photoionizing feedback and the resulting HII bubbles on the turbulent velocity fields in simulations of GMCs (developed by Dr Jim Dale). In order to pin down the effect of photoionization, we have also analyzed control runs that do not include this effect (see snapshots from the simulations on the left). In the control runs, the velocity distribution shows signs of gravitational collapse and dissipation of energy and the initial turbulent velocity field is erased. By contrast, in the runs including photoionization, we find that GMCs whose morphology is strongly affected by photoionization (i.e. which develop large bubbles of HII) also show evidence of driving of turbulence. The details can be found in Boneberg et al. 2015