The DustPedia Archive is now available here!

A Brief History of Dust
Ant Jones, Institut d’Astrophysique Spatiale, Orsay, France.

This very brief history of dust is focused on the nature of the large interstellar grain population, i.e., grains with sizes of the order of a hundred to a few hundreds of nano-metres, than it is on particles two orders of magnitude smaller or a million times less massive (literally, nano-particles, i.e., particles with dimensions of the order of a nano-metre). Thus, a review of the history of the interstellar “polycyclic aromatic hydrocarbon” (PAH) hypothesis is not included here and PAHs are only mentioned in passing.

The beginnings

Interstellar dust has been an ongoing topic of study in astronomy for almost 90 years and yet we still do not fully understand what it is made of and how it interacts with the gas in the interstellar medium. Nevertheless, we have learned a lot about dust since the earliest studies. These began with the observation that distant stars are not as bright as they should be, which was inferred to be a result of the absorption and scattering of starlight by small solid particles in the intervening interstellar medium (ISM), an effect called interstellar reddening or selective absorption. The discovery of reddening is often attributed to Robert J. Trumpler for his study of the effect along lines of sight towards stars in open clusters (Trumpler 1930). However, the reddening effect was independently described by Carl Schalén (1929, 1931) and who is known to have visited Trumpler at the Lick Observatory before his work on reddening was published. Unfortunately, Schalén’s contribution to the study of reddening in the early 1930s appears to have been largely forgotten (Hockey et al. 2014).

By the mid 1930’s it was well-accepted that interstellar clouds of gas and dust dimmed and reddened the colours of distant stars. It was during this era that Jesse Greenstein undertook the difficult task of photographically measuring highly reddened B stars, showing that the dust absorption followed an almost universal λ-0.7 dependence (Kraft 2005). The modelling of stellar reddening by dust began soon after Trumpler and Schalén’s interstellar absorption studies. Following up on his observations Greenstein used Gustave Mie’s theory (Mie 1908) to show that the observed wavelength dependence of the reddening could be reproduced by an appropriate size distribution of water ice, silicate or metal particles (Kraft 2005). However, similar dust modelling studies were more rigorously pursued by others. Indeed, Schalén (1934) and Schoenberg and Jung (1934) also used Mie’s theory to study the diffusion of starlight by interstellar matter and found that the observed interstellar absorption could be reproduced by metallic particles. We should perhaps recall that at the time calculations with Mie’s theory were difficult and time-consuming and so these early modelling attempts were indeed truly pioneering.

Some ten years after the earliest attempts at modelling interstellar reddening, the so-called “dirty ice” model for dust was proposed by Hendrik van de Hulst (1943) and then further developed by Jan Oort & Hendrik van de Hulst (1946). In the latter work Oort and van de Hulst determined that a dirty ice particle size distribution could be maintained through the effects of accretion from the gas, on a time-scale of ≈108 yr, and that this would be balanced by sublimation and evaporation in grain collisions, resulting in grains with average radii of the order of 150 nm and life-times of ≈ 50 million years. Thus, Oort and van de Hulst laid the groundwork for all later studies of interstellar dust evolution, which has been an important consideration since the first dust models were developed. A reading of both of their ground-breaking papers is therefore highly recommended to the reader.

The graphite and silicate dust model

Almost two decades after van de Hulst and Oort proposed the dirty ice model, Fred Hoyle and Chandra Wickramasinghe suggested that grains comprised of small graphite flakes, assumed to be formed around cool carbon stars, could provide an alternative explanation for the visible to near-IR interstellar extinction (Hoyle and Wickramasinghe 1962). Thus, with the clear advantage of hindsight, one might perhaps consider that Hoyle and Wickramasinghe were some 20 years ahead of their time in predicting the presence of PAHs/graphene platelets in the ISM; a hypothesis that has still to be convincingly proven. Soon after the suggestion of interstellar graphite flakes Wickramasinghe showed that silicate and graphite grains, with or without ice mantles, were consistent with the observed extinction (Wickramasinghe 1963, 1970). Then, with the measurement of the interstellar ultraviolet (UV) extinction and the discovery of a broad bump in this wavelength range at ∼217 nm by Ted Stecher (Stecher 1965, 1969), graphite became the favoured dust material and a likely carrier for the so-called UV extinction bump (Stecher & Donn 1965). At about the same time Hoyle and Wickramasinghe (1969) proposed that graphite along with silicate grains, formed around oxygen-rich giant stars, were likely to be the major interstellar dust species. These works, based on laboratory-measured optical properties for dust analogue materials, thus laid the foundations of the interstellar graphite and silicate dust models that followed. After almost fifty years these materials still remain the basis of almost all current interstellar dust models.

In the late seventies John Mathis and his co-researchers further refined the graphite/silicate dust model and, using laboratory-measured optical constants, explored the viability of un-coated graphite, enstatite, olivine SiC, iron and magnetite as suitable interstellar dust analogues (Mathis et al. 1977, hereafter MRN). From their study Mathis et al. concluded that a viable model for interstellar extinction must include graphite and that, provided it is combined with any of the other materials in a power law grain size distribution, gives a good fit to the observed near-infrared (NIR) to UV extinction. The preferred MRN dust model was for graphite grains partnered with an olivine- type silicate. Thus reinforcing graphite’s reputation as an essential interstellar dust component and confirming the graphite and silicate mix as the dust model of choice.

Basing their work upon the earlier MRN dust model Bruce Draine and Hyung Mok Lee (1984) semi-empirically tuned the optical properties of graphite and amorphous silicate materials to fit observations, resulting in what have become widely known as “astronomical” graphite and silicate interstellar dusts. PAH molecules were subsequently added into the “astronomical” graphite and silicate dust mix (Desert et al. 1990; Siebenmorgen & Kruegel 1992; Dwek et al. 1997; Draine & Li 2001, 2007; Li & Draine 2001, 2002; Siebenmorgen et al. 2014) but the physical basis for these models, even including the later and major improvements, was fundamentally little different from the original MRN model. It should be noted that in all of these models the different grain populations, for example, graphite and olivine (MRN), astronomical graphite and silicate (Draine & Lee 1984) or astronomical graphite, silicate and PAH (Siebenmorgen & Kruegel 1992; Dwek et al. 1997; Draine & Li 2001, 2007; Li & Draine 2001, 2002; Siebenmorgen et al. 2014) are considered to reside in distinct and separate dust populations. Some later dust models followed this same basic approach but abandoned astronomical graphite in favour of physically more-realistic amorphous carbons (e.g., Zubko et al. 2004; Compiègne et al. 2011; Galliano et al. 2011). However, most of these models suffer the same limitations as all MRN-based models, that is “The particles responsible for the 2200 Å hump and the FUV extinction constitute two independent populations, and the MRN graphite plus silicate model violates this condition.” (Greenberg & Chlewicki 1983) and “The ultraviolet extinction by the particles responsible for the visual extinction is grey and therefore these particles constitute an entirely separate population from the hump particles or the FUV particles.” (Greenberg & Chlewicki 1983). In contrast, the Desert et al. (1990) dust model in the InfraRed Astronomical Satellite (IRAS) era was built to be consistent with the Greenberg & Chlewicki (1983) constraints and was something of a departure from previous models in that it was the first of these new dust models to introduce PAHs into the mix. This model also assumed that the small grains are made of an amorphous carbonaceous material and that the large silicate grains have a “dark refractory mantle” of hydrocarbon composition.

Towards a complex realism

Given the turbulent nature of the ISM and the observation that the depletions of the dust-forming elements vary significantly and differentially (e.g., Routly & Spitzer 1952; Crinklaw et al. 1994; Savage & Sembach 1996; Jones 2000; Welty et al. 2002; Jenkins 2009; Parvathi et al. 2012) it is difficult to see how the two principal dust materials (carbon and silicate) could remain distinct and separate. The cycling of gas and dust through different ISM phases naturally requires that the dust materials become mixed into inhomogeneous assemblages. This mixing could be in the form of graphite and silicate core ice-mantled grains (Wickramasinghe 1963) or refractory-mantled particles, along the lines of those originally proposed by Greenberg (1986), in porous particles or mixed-material aggregates (e.g., Jones 1988; Mathis & Whiffen 1989). Indeed, physically-viable interstellar dust models based on silicate grains with “organic” or (hydrogenated) amorphous carbon mantles were first discussed by Mayo Greenberg (Greenberg 1986) and so have, in fact, been around for more than 30 years (e.g., Greenberg 1986; Duley 1987; Jones et al. 1987, 1990, 2013, 2014; Duley et al. 1989; Jones 1990; Sorrell 1991; Li & Greenberg 1997; Iatì et al. 2008; Cecchi- Pestellini et al. 2010; Zonca et al. 2011; Cecchi-Pestellini et al. 2014a,b; Köhler et al. 2014). As per all dust models, variations in the grain size distribution can be used to explain the observed variations in dust emission and extinction. However, a major advantage of the core-mantle models is that they can, in addition and quite naturally, account for dust variations through environmentally- driven changes in the carbonaceous mantle composition and/or its depth (e.g., Jones et al. 1987, 1990, 2013, 2016; Köhler et al. 2014; Duley et al. 1989; Li & Greenberg 1997; Cecchi-Pestellini et al. 2010; Zonca et al. 2011; Cecchi-Pestellini et al. 2014a,b; Ysard et al. 2015, 2016).

Dust evolution

It is thus clear that we can no longer think of dust as being the same everywhere, made of the same materials everywhere, nor of consisting of separated populations of distinct composition. Thus, dust modelling studies and viable dust models must now consider the nature of the dust properties as they evolve in response to the physical conditions in their local environment, e.g., density, radiation field (intensity and hardness as effected by extinction) and kinematics/dynamics. This evolution drives the dust properties (structure and composition) towards a state that is assumed to be in equilibrium with the given local conditions. For example, in diffuse and dense clouds, where the dust seemingly resides undisturbed for millions of years, the typical dust processing time-scales are short with respect to the dynamical time-scales and the dust is generally in equilibrium with the local conditions in these regions. However, in photon-dominated regions (PDRs) and shocked regions this is no longer the case because the dynamical time-scales are significantly shorter and the dust is often out of equilibrium and therefore constantly evolving in response to the changing local physical conditions.

The long-standing and widely-adopted idea of core-mantle (CM) interstellar grains (e.g., Greenberg 1986; Duley 1987; Jones et al. 1987, 1990; Duley et al. 1989; Jones 1990; Sorrell 1991; Li & Greenberg 1997; Iatì et al. 2008; Cecchi-Pestellini et al. 2010; Zonca et al. 2011; Cecchi-Pestellini et al. 2014a,b) was recently given a new treatment (Jones et al. 2013, 2017; Köhler et al. 2014). The result of this is the THEMIS dust modelling framework (see next section), which provides a core- mantle model for dust in the diffuse ISM and the evolution of the dust properties in response to their local environment (Jones et al. 2013, 2014, 2016; Köhler et al. 2014; Bocchio et al. 2014; Ysard et al. 2015, 2016). Indeed, variations in the dust properties from one region to another have long been interpreted as due to dust evolution (e.g., Lefevre 1974; Jura 1980; Aannestad & Greenberg 1983; Whittet et al. 1992, 2001; Ossenkopf 1993; Stepnik et al. 2003). For example, photon, ion and electron irradiation can induce changes in the dust chemical composition and structure (e.g., Demyk et al. 2001, 2004), hydrogenation and accretion can drive changes in the grain chemical composition (e.g., Hecht 1986; Sorrell 1990, 1991; Mennella 2008, 2010; Jones 2012a) and accretion/coagulation will change the grain structure (e.g., Köhler et al. 2011, 2012, 2015; Jones et al. 2014, 2016). All of these processes directly affect the dust optical properties (i.e., the particle absorption and scattering cross-sections) as the composition, structure and shape of the grains evolve in the transition between regions with different physical conditions. For example, in the tenuous ISM the outer carbonaceous layers of the grains, be they carbon grains or the mantles on other grains, will be H-poor and aromatic rich due to UV photolysis by stellar FUV/EUV photons (e.g., Hecht 1986; Sorrell 1990, 1991; Jones 2012a,b,c, 2016; Jones et al. 2013) but in denser regions accreted hydrocarbon mantles are likely H-rich and aromatic poor (Jones et al. 2014, 2016; Ysard et al. 2016). The accretion of these H-rich mantles does indeed appear to be entirely consistent with observations of cloud- and core-shine in the outer regions of molecular clouds (Jones et al. 2016; Ysard et al. 2016; Togi et al. 2017) and in the transition from diffuse to dense clouds (Köhler et al. 2015).

The THEMIS framework

The underlying principle of the recently-proposed evolutionary dust modelling approach, THEMIS (The Heterogeneous dust Evolution Model for Interstellar Solids: Jones et al. 2013, 2016, 2017; Köhler et al. 2015; Ysard et al. 2015, 2016), is the supposition that interstellar dust is not the same everywhere but that it evolves within a given region of the ISM as it reacts to and interacts with its local environment. Thus, THEMIS is in fact not a single dust model but a framework that coherently considers the evolution of the dust properties within given regions and as the dust transits between different regions of the ISM. THEMIS considers the evolution of this dust in response to the physical conditions in the ambient medium. The THEMIS modelling approach is global in the sense that the observable dust properties are considered as a whole, from extreme UV to cm wavelengths, including self-consistent spectroscopic properties, and also in the sense that it can be applied to a wide variety of interstellar environments. This modelling approach provides a new view of interstellar dust and a useful framework within which many aspects of dust and its evolution in the ISM can be explored and tested.

THEMIS is built around a core diffuse ISM dust model that, for the first time, includes size- dependent carbonaceous dust optical properties, something that does not yet exist for silicate or other materials but that does need to be included in future models.

Thoughts for the future

Recent observations have shown that the nature of interstellar dust is inherently and significantly much more complex than had previously been considered. New physically-realistic dust modelling approaches need to be stoutly anchored to the laboratory-measured properties of cosmic dust analogues, a methodology that leaves little room for arbitrary observation-fitting adjustments. A global, laboratory data-constrained approach appears to be an inherently more successful approach than models that empirically adjust dust properties to fit the observations. While empirically-based dust models are undoubtedly useful, in that they can be finely tuned to the observations and thereby advance astrophysical understanding, in achieving this they are unable to significantly advance our understanding of the micro- and nano-physical properties of dust.

It is clear that we need to rigidly constrain future dust models using laboratory-measured properties and to understand how the properties of these interstellar dust analogue materials are likely to evolve under the varied physical conditions within interstellar media.


This history is necessarily incomplete and also somewhat unavoidably biased by the author’s particular viewpoint. Hopefully it will tend towards completeness with further additions, deletions and modifications. The author wishes to thank Derek Ward-Thomson for the suggestion to publish this brief history. Nevertheless, the author accepts full responsibility for all errors, omissions and inaccuracies.


References Aannestad, P. A., & Greenberg, J. M. 1983, ApJ, 272, 551.
Bocchio, M., Jones, A. P., & Slavin, J. D. 2014, A&A, 570, A32.
Cecchi-Pestellini, C., Cacciola, A., Iatì, M. A., et al. 2010, MNRAS, 408, 535.
Cecchi-Pestellini, C., Casu, S., Mulas, G., & Zonca, A. 2014a, ApJ, 785, 41.
Cecchi-Pestellini, C., Viti, S., & Williams, D. A. 2014b, ApJ, 788, 100.
Compiègne, M., Verstraete, L., Jones, A., et al. 2011, A&A, 525, A103.
Crinklaw, G., Federman, S. R., & Joseph, C. L. 1994, ApJ, 424, 748.
Demyk, K., Carrez, P., Leroux, H., et al. 2001, A&A, 368, L38.
Demyk, K., d’Hendecourt, L., Leroux, H., Jones, A. P., & Borg, J. 2004, A&A, 420, 233. Desert, F.-X., Boulanger, F., & Puget, J. L. 1990, A&A, 237, 215.
Draine, B. T., & Lee, H. M. 1984, ApJ, 285, 89.
Draine, B. T., & Li, A. 2001, ApJ, 551, 807.
Draine, B. T., & Li, A. 2007, ApJ, 657, 810.
Duley, W. W. 1987, MNRAS, 229, 203.
Duley, W. W., Jones, A. P., & Williams, D. A. 1989, MNRAS, 236, 709.
Dwek, E., Arendt, R. G., Fixsen, D. J., et al. 1997, ApJ, 475, 565.
Galliano, F., Hony, S., Bernard, J.-P., et al. 2011, A&A, 536, A88.
Greenberg, J. M. 1986, Ap&SS, 128, 17.
Greenberg, J. M., & Chlewicki, G. 1983, ApJ, 272, 563.
Hecht, J. H. 1986, ApJ, 305, 817.
Hockey, T., Trimble, V., Williams, T. R., et al. 2014, Biographical Encyclopedia of Astronomers (Berlin: Springer).
Hoyle, F., & Wickramasinghe, N. C. 1962, MNRAS, 124, 417.
Hoyle, F., & Wickramasinghe, N. C. 1969, Nature, 223, 459.
Iatì, M. A., Saija, R., Borghese, F., et al. 2008, MNRAS, 384, 591.
Jenkins, E. B. 2009, ApJ, 700, 1299.
Jones, A. P. 1988, MNRAS, 234, 209. Jones, A. P. 1990, MNRAS, 247, 305.
Jones, A. P. 2000, J. Geophys. Res., 105, 10257.
Jones, A. P. 2012a, A&A, 540, A1.
Jones, A. P. 2012b, A&A, 540, A2; Corrigendum: 545, C2.
Jones, A. P. 2012c, A&A, 542, A98; Corrigendum: 545, C3.
Jones, A. P. 2016, R. Soc. Open Sci., 3, 160224
Jones, A. P., Williams, D. A., & Duley, W. W. 1987, MNRAS, 229, 213.
Jones, A. P., Duley, W. W., & Williams, D. A. 1990, QJRAS, 31, 567.
Jones, A. P., Fanciullo, L., Köhler, M., et al. 2013, A&A, 558, A62.
Jones, A. P., Ysard, N., Köhler, M., et al. 2014, RSC Faraday Discuss., 168, 313.
Jones, A. P., Köhler, M., Ysard, N., et al. 2016, A&A, 588, A43.
Jones, A. P., Köhler, M., Ysard, N., Bocchio, M., Verstraete, L. 2017, A&A, 602, A46.
Jura, M. 1980, ApJ, 235, 63.Kraft, R. P. 2005, Biographical Memoirs, The National Academies Press, 86, 162.
Köhler, M., Guillet, V., & Jones, A. 2011, A&A, 528, A96.
Köhler, M., Stepnik, B., Jones, A. P., et al. 2012, A&A, 548, A61.
Köhler, M., Jones, A., & Ysard, N. 2014, A&A, 565, L9.
Köhler, M., Ysard, N., & Jones, A. P. 2015, A&A, 579, A15.
Lefevre, J. 1974, A&A, 37, 17.
Li, A., & Draine, B. T. 2001, ApJ, 554, 778.
Li, A., & Draine, B. T. 2002, ApJ, 576, 762.
Li, A., & Greenberg, J. M. 1997, A&A, 323, 566.
Mathis, J. S., & Whiffen, G. 1989, ApJ, 341, 808.
Mathis, J. S., Rumpl, W., & Nordsieck, K. H. 1977, ApJ, 217, 425 (MRN).
Mennella, V. 2008, ApJ, 682, L101.
Mennella, V. 2010, ApJ, 718, 867.
Mie, G. 1908, Annalen der Physik, 330, 377.
Oort, J. H., & van de Hulst, H. C. 1946, Bull. Ast. Inst. Netherlands, 10, 187.Ossenkopf, V. 1993, A&A, 280, 617.
Parvathi, V. S., Sofia, U. J., Murthy, J., & Babu, B. R. S. 2012, ApJ, 760, 36.
Routly, P. M., & Spitzer, Jr., L. 1952, ApJ, 115, 227.
Savage, B. D., & Sembach, K. R. 1996, ARA&A, 34, 279.
Schalén, C. 1929, Astron. Nachr., 236, 249.Schalén, C. 1931, Publ. Am. Astron. Soc., 6, 376.
Schalén, C. 1934, Meddel. Astron. Obs. Upsala, 58. Schoenberg, E., & Jung, B. 1934, Astron. Nachr., 253, 261.
Siebenmorgen, R., & Kruegel, E. 1992, A&A, 259, 614.
Siebenmorgen, R., Voshchinnikov, N. V., & Bagnulo, S. 2014, A&A, 561, A82.
Sorrell, W. H. 1990, MNRAS, 243, 570.
Sorrell, W. H. 1991, MNRAS, 248, 439.
Stecher, T. P. 1965, ApJ, 142, 1683.
Stecher, T. P. 1969, ApJ, 157, L125.
Stecher, T. P., & Donn, B. 1965, ApJ, 142, 1681.
Stepnik, B., Abergel, A., Bernard, J.-P., et al. 2003, A&A, 398, 551.
Togi, A., Witt, A.N., St. John, D. 2017, A&A, in press.
Trumpler, R. J. 1930, PASP, 42, 214.
van de Hulst, H. C. 1943, Ned. Tijdschr. V. Natuur, 10, 251.
Welty, D. E., Jenkins, E. B., Raymond, J. C., Mallouris, C., & York, D. G. 2002, ApJ, 579, 304.
Whittet, D. C. B., Martin, P. G., Hough, J. H., et al. 1992, ApJ, 386, 562.
Whittet, D. C. B., Gerakines, P. A., Hough, J. H., & Shenoy, S. S. 2001, ApJ, 547, 872. Wickramasinghe, N. C. 1963, MNRAS, 126, 99.
Wickramasinghe, N. C. 1970, in Ultraviolet Stellar Spectra and Related Ground- Based Observations, eds. R. Muller, L. Houziaux, & H. E. Butler, IAU Symp., 36, 42.
Ysard, N., Köhler, M., Jones, A., et al. 2015, A&A, 577, A110.
Ysard, N., Köhler, M., Jones, A., et al. 2016, A&A, 588, A44.
Zonca, A., Cecchi-Pestellini, C., Mulas, G., & Malloci, G. 2011, MNRAS, 410, 1932. Zubko, V., Dwek, E., & Arendt, R. G. 2004, ApJS, 152, 211.