Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton. 16 January 1923 −− 29
May 2007
Philip G. Burke and Peter J. Storey
Biogr. Mems Fell. R. Soc. published online February 13, 2013
Supplementary data
"Data Supplement"
http://rsbm.royalsocietypublishing.org/content/suppl/2013/02
/12/rsbm.2012.0044.DC1
P<P
Published online 13 February 2013 in advance of the print
journal.
Email alerting service
Receive free email alerts when new articles cite this article sign up in the box at the top right-hand corner of the article
or click here
Advance online articles have been peer reviewed and accepted for publication but
have not yet appeared in the paper journal (edited, typeset versions may be posted
when available prior to final publication). Advance online articles are citable and
establish publication priority; they are indexed by PubMed from initial publication.
Citations to Advance online articles must include the digital object identifier (DOIs)
and date of initial publication.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
16 January 1923 — 29 May 2007
Biogr. Mems Fell. R. Soc.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
16 January 1923 — 29 May 2007
elected FRS 1967
by philip g. burkE1 cbE frs and pEtEr j. storEy2
1Department
of Applied Mathematics and Theoretical Physics,
Queen’s University Belfast, Belfast BT7 1NN, UK
2Department of Physics and Astronomy, University College London,
London WC1E 6BT, UK
Mike Seaton was an outstanding atomic physicist and astrophysicist with a wide range of
research interests. his work at University college london (Ucl), where he spent all his professional career, laid the foundations of the modern theory of electron–atom and electron–ion
collisions and the crucial role of this theory and associated calculations in the analysis of
astronomical spectra. his work in astrophysics ranged from seminal papers on the central stars
of planetary nebulae and density diagnostics, using forbidden lines, to the Seaton extinction
curve, the theory of diffusion in stars and many other topics. indeed, in the 1950s and 1960s
Mike was publishing papers on a whole range of topics that each became major research areas
in the years that followed. For example, Mike’s collaboration over some 30 years with Donald
osterbrock laid the foundation of our understanding of physical processes in planetary nebulae. in addition, while on sabbatical at the institut d’astrophysique de Paris in 195–55 Mike
began a highly influential series of papers on quantum defect theory (QDt); applications of
this theory were to resonance phenomena in electron–ion collisions and to photoionization. in
addition, Mike led major research programmes on radiative recombination in gaseous nebulae
and on dielectronic recombination. Mike and his group at Ucl were leaders in the development of general computer codes for atomic structure and collision processes. this made
possible the first accurate electron–Fe ii calculation. Major initiatives that grew out of the
availability of these computer codes were the international opacity Project, proposed and led
by Mike, and the international iron Project proposed by and initially led by David hummer.
these projects, which involved collaboration with research workers from many countries,
have been of crucial importance, solving many problems in atomic physics and astronomy.
Mike Seaton was President of the Royal astronomical Society for the period 1978–81, and
http://dx.doi.org/10.1098/rsbm.2012.00
3
© 2013 the author(s)
Published by the Royal Society
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Biographical Memoirs
his tenure at UCL coincided with the golden age of atomic astrophysics, for which he was
largely responsible.
Early life and education
Michael John Seaton was born on 16 January 1923 at 4 Arlington Villas, Clifton, Bristol,
the son of Arthur William Robert Seaton (1892–1959) and his wife Helen Amelia, née Stone
(1893–1985). His father was a Baptist lay preacher in his youth and a stretcher-bearer in World
War I. Later he was a commercial traveller for a paper manufacturer. His mother was a nurse in
World War I, and after the war she went to Serbia with the first Quaker relief nursing team. In
the 1930s she joined the Communist Party but she ended her life as a follower of the Quakers.
As well as Mike, his parents had two other children: an elder daughter, Pauline Joan, and a
younger son, Colin Robert.
At the age of about 5 or 6 years Mike was struck down with rheumatic fever and was
effectively an invalid for the next few years. Subsequently at school he was always a year
or two older than the rest of the class. His first interest in science was when his sister was
given a chemistry set by Father Christmas at Selfridges grotto. This was a subject she hated,
and Mike’s gift was snatched out of his hands and exchanged for the chemistry set. This set
enthralled Mike and he did a lot of experiments, mainly in inorganic quantitative analysis and
organic preparation, in a shed in the garden. Unfortunately this was completely destroyed in
an unexpected explosion, but Mike escaped unscathed.
In his teens Mike and his friends went on extensive cycling trips in Britain and France. He
continued exploring by bicycle for many years. He was also a keen swimmer.
At the age of 15 years, in 1938, Mike was horrified to hear Chamberlain’s speech broadcast
on his return from Munich and, as a result, joined the Young Communist League, although
politics took second place to chemistry. In 1940 he was arrested at a street-corner meeting
and was convicted of ‘using threatening, insulting and abusive language’, which he had not
used. As a result he was expelled from school (Wallington County Grammar School, Surrey)
just before the matriculation exams. However, he was allowed to take the exams and obtained
creditable marks in all subjects. Years later he was invited back to the school to present the
end-of-year prizes, the irony of which pleased him greatly. His affiliation to the Communist
Party ended in 1956 after the Soviet invasion of Hungary.
World War II
After leaving school in 1940, Mike worked for two years in a chemical factory, British
Industrial Solvents, doing routine analysis of their products. He then volunteered for Air
Crew, although he was uncertain whether he would pass the medical examination as a result
of his childhood rheumatic fever. However, he did and was called up in 1942. His first posting
was to the Initial Training Wing in Cambridge, where he was billeted in Pembroke College.
While there he met his first wife, Olive May Singleton, who was then serving in the Women’s
Auxiliary Air Service. She was the daughter of Charles Singleton, and before her war service
she had worked as a shorthand typist. They were married on 10 June 1943; their marriage
lasted nearly 16 years until her early death in 1959. They had a son, Richard, and a daughter,
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
Jane. Almost immediately after the wedding Mike was posted to Winnipeg to complete his
training and during a 10-day leave paid his first visit to the USA. In the spring of 1944, after
graduating as an air navigator, he was given a further 10 days’ leave and, with several companions, paid his second visit to the USA, which he spent seeing the sights in New York.
Mike returned from Canada to the UK in 1944 as a Flight Lieutenant and was appointed
Navigator in the Pathfinder Force of Bomber Command flying Lancaster bombers. Altogether
he flew about 20 missions over Europe with only one really nasty experience, when much of
the underside of one wing was ripped off by anti-aircraft fire. They limped home to a ground
crew who were very cross that they had not taken better care of their plane. However, there
were numerous stories of how Mike’s skill with instruments and the precision of his calculations, later to be so evident in his research, saved his crew from enemy action. On one occasion, shortly after the end of the war, he was flying in a Lancaster Bomber to Bari in southern
Italy to bring home prisoners of war and refugees. The crew decided to make an unscheduled
detour to see Mont Blanc. They flew through very heavy cloud and were dismayed to suddenly
see the peak of Mont Blanc extremely close. It took all Mike’s ingenuity and computational
skill as navigator to avoid a crash.
University College London
After the end of World War II Mike enrolled as an external student at London University, and
by the time of his discharge from the RAF in 1946 he had passed the Intermediate Science
Examination and had completed about two-thirds of the work for a general degree in physics
and mathematics. This was sufficient for him to be accepted as an undergraduate student in
physics at UCL.
Thereafter Mike spent all of his professional career at UCL apart from several visits abroad.
This section is thus the story of his life and his very many achievements in atomic physics
and astrophysics. We have selected four of his most notable contributions and describe them
in more detail in later separate sections.
Mike graduated with a first-class honours BSc degree in physics in 1948. The Physics
Department was then headed by Professor E. N. da C. Andrade FRS. To pursue research in
quantum theory, not then taught in the Physics Department, Mike moved to the Mathematics
Department, headed by Harrie (later Sir Harrie) Massey FRS (SecRS 1969–78). The period
immediately after the war marked the start of a new era, both for atomic physics and astrophysics at UCL and for the influence that UCL was to have internationally in these fields.
Having worked in the Admiralty and on the Manhattan Project in the USA during the war,
Massey had just returned to his appointment as Goldsmid Professor of Mathematics at UCL,
where he was joined by several colleagues including David (later Sir David) Bates (FRS
1955). Bates had been a research student working with Massey at Queen’s University Belfast
before the war and was now appointed Lecturer in Mathematics at UCL.
Mike began his research supervised by Bates in the Mathematics Department. Bates took
his role as supervisor very seriously and on one occasion gave a formal series of lectures that
only one student, Mike, attended. This clearly provides a strong argument against the undesirability of small classes. Between 1949 and 1951, during the course of work for his PhD thesis
entitled ‘Quantal calculations of certain reaction rates with applications to astrophysical and
geophysical problems’, Mike published six papers; significantly, they were equally distributed
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Biographical Memoirs
between Monthly Notices of the Royal Astronomical Society and physics journals. The theme
of his thesis was to become the basis of his subsequent scientific work.
In October 1950 Massey moved to the Physics Department at UCL to take up a new appointment as Quain Professor of Physics. He was accompanied by Bates, R. A. Buckingham and
E. H. S. Burhop (FRS 1963), who became readers in physics, and by Mike and L. Castillejo,
who became assistant lecturers. At about the same time Mike started writing his classic 1953
paper on the Hartree–Fock equations (1)*, which extended Hartree–Fock methods for atomic
structure calculations to the study of electron–atom collision calculations with application to
electron impact excitation of the ground-state configuration terms of atomic oxygen. As Mike
pointed out, although Hartree–Fock wavefunctions had previously been used in the study of
various continuum state problems, no complete justification for the procedure had been given
and the equations used, derived largely by analogy with bound-state equations, did not include
the coupling terms required in collision theory. The calculations on atomic oxygen were rapidly
extended by Mike to O ii, O iii and N ii (2) providing for the first time estimates of electron
impact excitation and de-excitation of the forbidden lines of the ground-state configurations for
these systems. These results and subsequent work by Mike were of fundamental importance
in many applications, including in the physics of the upper atmosphere and gaseous nebulae.
Mike’s research on forbidden lines in aurorae was reviewed by Bates (1983), who observed
that the research was conducted over quite a short period, 1953–58, but formed a key development in this subject. It included detailed calculations of electron impact excitation of forbidden
atomic lines in auroral spectra (4). In addition, Mike’s work with Osterbrock on O ii (5), using
the O ii doublet to measure electron densities in gaseous nebulae, was part of a lifelong collab­
oration between Mike and Osterbrock (figure 1), which was reviewed by Ferland (2008).
In 1954 Mike took leave of absence from UCL for the academic year 1954–55 and
went to France as Chargé de Recherche at the Institut d’astrophysique de Paris. This visit
was instigated by Mike’s need to work for a while in an institute where the interests were
purely astronomical, and it marked the start not only of his long association with H. van
Regemorter but also of the close collaboration that subsequently developed between UCL
and the Observatoire de Paris at Meudon. Later this collaboration was extended to include the
Observatoire de Nice. Also in 1954 Mike realized that QDT provided an important relationship between bound and continuum states of atomic systems, and his first publication on this
subject appeared in Comptes Rendus in 1955 (3). We review Mike’s role in the development
and application of QDT below.
On his return from France in 1956 Mike, in collaboration with I. C. Percival (FRS 1985),
initiated the first rigorous analysis of the partial wave theory of electron collisions with
hydrogen atoms, which was published in 1957 (6). In addition, one of us (P.G.B.) owes his
introduction to atomic collision physics to Mike in 1956 when as a new postdoctoral worker at
UCL he collaborated with Mike in the development of a computer code to solve these collision
problems. In 1960 Mike, in collaboration with L. Castillejo and Percival, analysed the longrange interactions between electrons and hydrogen atoms (9). In the same year he published
a review on planetary nebulae (10) that was quoted for many years subsequently. In the next
six years Mike collaborated with David Hummer and Reginald Harman on a series of papers
on the ionization structure of planetary nebulae, which laid out the theoretical and numerical
basis for the construction of models of these objects. The newly available ­ computers made
*Numbers in this form refer to the bibliography at the end of the text.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
Figure 1. Donald Osterbrock and Mike, with Joy Seaton (on the left) and Don’s wife, Irene. The photograph was taken
at a Baltimore Symposium in May 1994 in honour of the 70th birthdays of Mike and Don. (Online version in
colour.)
this a very active area worldwide in the 1960s. Mike’s important 1964 paper with Harman
(12) used observations of planetary nebulae to trace the track of their central stars on the
Hertzsprung–Russell diagram for the first time. Mike’s work on planetary nebulae led him
to discuss the inadequacies in the theory of recombination processes and to initiate a major
analysis of recombination processes, which we discuss below in a separate section.
Very tragically, early in 1959 Mike’s wife Olive died from cancer; nearly two years later
Mike married Joy Clarice Balchin on 20 October 1960. She was Sir Harrie Massey’s secretary
at UCL and is a gifted artist and lay preacher. She is the daughter of Harry Albert Balchin and
Clarice May Balchin. She took over the care of the children Jane and Richard, and in 1963
Joy and Mike had a son, Anthony. They lived in South London, where they were visited by
Mike’s colleagues from all over the world. There was also a constant stream of students who
enjoyed their hospitality.
In 1961 Mike and the family visited the University of Colorado at Boulder, USA for five
months when the Joint Institution for Laboratory Astrophysics (JILA) was being established,
and Mike became a founder member. While there, Mike attended the 2nd International
Conference on the Physics of Electronic and Atomic Collisions, chairing the session on electron–ion collisions. Later, in 1964, Mike was appointed a Fellow-Adjoint of the institute in
recognition of his outstanding contributions to its scientific activities. There followed an active
collaboration between UCL and JILA in both atomic physics and astrophysics. For several
years Mike made further visits to Boulder, sometimes with the family, who grew to love the
people and the country.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Biographical Memoirs
The increasing interest in atomic structure and electron–atom collisions resulted in intensive efforts by workers worldwide to solve the corresponding coupled integro-differential
equations that arise in these processes. In addition, with the rapidly increasing power and
availability of high-performance computers, several computer codes were developed to solve
these equations. The development of methods for this and to design and write the corresponding computer codes was later reviewed by Burke and Seaton (17). The role and importance
of high-performance computers in the solution of problems in atomic and molecular physics
were central to Mike’s research in this field.
Mike’s expertise in computing was recognized by his appointment as a member of the Flowers
Committee, which investigated the future needs of computing facilities in UK universities. Its
report, accepted by the government in 1966, resulted in major changes and led to the establishment of the Computer Board, setting the pattern for the next 20 years. He also served as Chairman
of the UCL Computer Board of Management for 10 years and was a member of the Board of
Management of the University of London Computer Centre and of the University’s Central
Coordinating Committee for Computing Services. His efforts in 1965 were largely responsible
for the installation of the IBM 360/65 computer at UCL. He also strongly supported the establishment of the North-Holland journal Computer Physics Communications (CPC) and the associated
program library in 1968, and he was a very active Advisory Editor for many years. It was through
Mike that UCL became one of the first subscribers to the CPC Program Library and maintained a
mirror of the library. He and his group contributed many important programs to it over the years.
We review the development of the UCL computer package in a separate section below.
An important area in which high-performance computing had a crucial role in Mike’s
research was the development of computer codes in the study of stellar opacities. In 1982,
during one of the annual visits that Mike made to Boulder, David Hummer brought him and
Dimitri Mihalas together to discuss a paper that Dimitri had read by Norman Simon which
suggested that there might be major errors in existing Rosseland mean opacity tables. The
result of these discussions was Mike’s initiating the International Opacity Project, which we
review below in a separate section. After the completion of the initial goals of the Opacity
Project, Mike was very gratified to find that the ex-students and collaborators who had worked
with him on that project were eager to continue using the skills acquired and to continue to
collaborate in the same spirit. As a result a new major collaborative project was proposed, and
initially led, by David Hummer, named the Iron Project, which we also review below. Under
the auspices of these two projects, Mike and the team members worked on commonly agreed
tasks and met twice a year for nearly two decades.
In 1988 Mike formally retired and was appointed Emeritus Professor at UCL on reaching
the age of 65 years. He continued working in his office there and collaborating with members
of staff and research students at UCL as well as with scientists worldwide. Indeed, his rate of
publication showed no diminution. We describe Mike’s final years in Wales in a later section,
followed by a reminiscence by Joy Seaton and two final tributes.
Quantum defect theory
QDT is concerned with the relationship between bound and continuum states of atomic
or molecular systems composed of an electron moving in the field of a positive atomic or
molecular ion. In particular, this theory expresses these properties in terms of analytic functions of energy. This provides a unified theory of bound states, including series perturbations,
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
autoionization and electron–ion collisions including radiative transition probabilities and
dielectronic recombination. In this section we discuss Mike’s crucial role in the development
and application of the theory.
The foundations of the modern QDT were laid by Hartree (1928). He considered the radial
Schrödinger equation describing electron–atom scattering,
 1  d 2 l (l + 1) 

[1]
−  2 −
 + V ( r ) − E  F ( E , r ) = 0 ,
2
r

 2  dr

in atomic units (e = m = ħ = 1), where V(r) is the potential due to the nucleus of charge Z0
screened by the spherically averaged charge distribution of N core electrons. For sufficiently
large radius r of the scattered electron there is complete screening and the potential V(r) = −Z/
r, where Z = Z0 − N. The quantum defect then represents the effect of the non-Coulomb part
of the potential and is zero when the non-Coulomb part of the potential vanishes. The solutions of equation [1] can be obtained in terms of Coulomb functions, and QDT hinges on their
mathematical properties, which were developed and programmed for computers by Mike.
These codes were later published as a comprehensive package (34–36).
Further interest in the theory was stimulated by the work of Bates & Damgaard (1949),
whose Coulomb approximation provided a powerful method for the computation of bound–
bound oscillator strengths for simple atomic systems. Finally, early work on QDT was discussed by Hartree (1957, chapter 8).
Building on this previous work, Mike played a crucial role in the development and application of QDT. After his early work on electron–ion collisions, he initiated a major programme
of research on QDT while on sabbatical leave in 1954–55 at the Institut d’astrophysique
de Paris. This resulted in a highly influential series of research papers on QDT (including
(3, 7, 15, 16, 22)), followed by a major review of the subject (25). As well as describing in
detail the foundations of the theory, that review discussed a range of applications of QDT,
including extension of the theory to multichannel quantum defect theory (MQDT). The review
also included detailed discussions of resonances, atomic collision calculations, systems with
two energy levels of the ion core, helium, other rare gases, alkaline earths and other atomic
systems, molecular hydrogen and dielectronic recombination.
Mike’s work on MQDT led to further applications by other workers. For example, Fano
(1970) extended and applied MQDT to the analysis of high-resolution photoabsorption spectra
of H2 near threshold. The work on H2 was later extended by Jungen & Atabek (1977) to rovibrational interactions in the photoabsorption spectrum of H2 and D2 and by Jungen & Dill
(1980), who studied rotational and vibrational preionization channels of H2. Finally, a review
of developments and applications of molecular MQDT has been written by Greene & Jungen
(1985). A general review of MQDT has also been written by Moores & Saraph (1983) with
emphasis on the crucial contributions of Mike Seaton.
Recombination processes
Radiative recombination in gaseous nebulae
The recombination lines of hydrogen are essential in determining elemental abundances relative to hydrogen and are prominent in the spectra of photoionized nebulae such as planetary
nebulae and H ii regions. Baker & Menzel (1938) calculated the expected strength of these
lines up to n = 20, accounting for the processes of radiative recombination and decay, but Aller
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
10
Biographical Memoirs
et al. (1955) reported photographic observations with the telescopes at Mount Wilson of the
high members of the Balmer series of hydrogen that were in serious disagreement with theory.
They attributed the disagreement to inadequacies in the theory, especially the neglect of collisional processes. Mike (11) examined these suggestions and gave arguments refuting them,
showing that a systematic overestimation of the intensities of weak lines was a more plausible
explanation; he called for accurate photometry of the weak lines, suggesting that the differences were due to calibration problems in the photographic observations. From a theoretical
perspective, the effect on the recombination spectrum of collisional processes between high-n
states of hydrogen and other hydrogenic ions was unexplored at that time. In the subsequent
decade, Mike and his students attacked and solved this problem.
Mike first dealt with the purely radiative recombination and cascade problem originally discussed by Baker & Menzel (1938), giving accurate formulae based on asymptotic expansions
for the radiative recombination coefficient to states of arbitrarily large n. He then presented a
solution of the capture–cascade equations for hydrogen, using a cascade matrix formalism that
enabled him to calculate the intensities of the high Balmer lines, correcting the work of Baker &
Menzel (1938) in the process. As the principal quantum number n increases, the rate of collisions
with free electrons increases as n4 while the total radiative decay rate falls as n−5, so that collisional processes eventually become dominant, irrespective of the free electron number density.
The largest collision rate is that between adjacent n states, and Mike showed in 1962 how to
calculate the rate for these transitions in a semi-classical approximation. This opened the way
for a treatment of the level populations of hydrogen for high values of n. Mike (14) showed how
the equilibrium equations for the level populations could be written as a second-order differential
equation in n and solved that equation to obtain the correct behaviour for the high-n populations.
This work was continued by M. Brocklehurst under Mike’s supervision and resulted in the publication of comprehensive tables of populations and population gradients (Brocklehurst 1970),
which were needed to interpret the newly discovered radio recombination lines of hydrogen.
This work culminated in a review article on the interpretation of the radio recombination lines
and continuum with particular emphasis on a model of the Orion Nebula (18).
The hydrogen population models used for the radio recombination lines characterized
each state by its principal quantum number n, tacitly assuming that population is rapidly
distributed among the nl states by collisional processes. This no longer holds for the low-n
states which give rise to optical recombination lines seen in nebulae, so a treatment of the
population problem in terms of nl rather than n was needed. Already in 1958 Burgess, then
working with Mike, had investigated the effect of resolving the nl states in the purely radiative approximation and found substantial differences from the earlier work of Baker & Menzel
(1938). Pengelly and Seaton (13) developed Mike’s semi-classical impact parameter method
to treat the collisions with ions, H+, He+ and He2+, which cause l-changing collisions, and
the results of a full treatment of the population problem for nl states of hydrogen was given
by Brocklehurst (1971). His calculations used a matrix condensation method to condense the
infinite set of equilibrium equations for the populations to a finite set soluble by simple matrix
inversion. It is noteworthy that although Mike wrote key sections of the computer code for
the condensation process used by Brocklehurst, he nonetheless encouraged him to be the sole
author on the paper, which remained the essential source of theoretical relative intensities for
optical recombination lines of hydrogen for 25 years.
The treatment of the recombination lines of hydrogen and hydrogenic ions can draw on
exact expressions for radiative transition probabilities and photoionization cross-sections for
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
11
these ions. Next to hydrogen, the spectrum of helium is the most important astronomically,
with the helium abundance having cosmological significance. In 1960 Burgess and Seaton
(8) presented a general formula for the calculation of photoionization cross-sections for nonhydrogenic states, using the quantum defect of the initial state and the continuum state phase
shift obtained by extrapolating bound-state quantum defects. This opened up the possibility
of obtaining more accurate recombination coefficients for low-l states for simple atoms but
especially for helium. Brocklehurst (1972) incorporated these techniques in a non-hydrogenic
version of his population code to provide definitive recombination coefficients for helium
lines, which remained the standard source for interpreting helium spectra until the 1990s.
Dielectronic recombination
Massey and Bates coined the term dielectronic recombination in 1942 to refer to recombination via a doubly-excited autoionizing state and Bates & Massey (1943) used detailed balance
arguments to obtain an expression for the rate coefficient for this process. Their interest was
in recombination in the ionosphere and estimates indicated that dielectronic recombination
would not be a significant process there. Mike tells (20) how, in 1961, Albrecht Unsold
­suggested to him privately that perhaps dielectronic recombination is of importance in the
ionization equilibrium of the solar corona, where measurements of the temperature from line
widths were substantially higher than those obtained from modelling the ionization balance. In
a brief paper, which Mike often spoke ruefully about in later years, he argued that there could
be no systematic increase in recombination due to the dielectronic process. Burgess (1964),
who was working in Mike’s group at the time, showed that Mike’s argument was not valid at
the high temperatures found in the solar corona, where the recombination occurs via Rydberg
series of resonance states; together they went on to show that dielectronic recombination is the
dominant recombination mechanism in the solar corona and that the temperature discrepancy
was resolved.
There followed a surge of activity by several authors attempting to put the theory of dielectronic recombination on a rigorous quantum-mechanical footing and to confirm that the intuitive
arguments underlying the work of Burgess were valid. His arguments, and those of Bates &
Massey (1943), treated the autoionizing states as isolated, but for Rydberg series of such states
the radiative width, which is dominated by a transition in the ion core, is essentially constant
and must eventually become larger than the separation of adjacent states. Bell and Seaton (26),
building on earlier work by Mike on dielectronic recombination and QDT, derived an expression
for the dielectronic recombination probability that is valid for the region where the resonance
states are no longer isolated. It differs from that used by Burgess but leads to results for the total
recombination coefficient that differ little from his, being only 3% different in a test case.
Further insights into the role of dielectronic recombination in astronomy were still to come;
they came, as is so often the case, from the opening up of a new spectral region, in this case
the ultraviolet. Mike (32) recounts how he inadvertently became an observational astronomer
with access to the International Ultraviolet Explorer (IUE) satellite when it was launched in
January 1978. He tells how Bob (later Sir Robert) Wilson (FRS 1975), who was the creator
and driving force behind IUE, asked him in about 1970 what objects he would like to observe
in the ultraviolet and how he had made a case to him for observing planetary nebulae and
novae. When IUE was launched, Mike was informed, much to his surprise, that his application
for observing time had been successful and he went on to observe planetary nebulae and also
became coordinator of the European Target of Opportunity Team for novae.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
12
Biographical Memoirs
When Nova Cygni 1978 erupted in September of that year, the ultraviolet spectra arrived at
UCL. Before they could be analysed, the effects of reddening by interstellar and circumstellar
dust had to be accounted for. Mike collated the data from previous observations and published
an ultraviolet extinction curve fitted to a simple, easy to use, algebraic expression (23), which
quickly became the standard form for de-reddening ultraviolet spectra. The spectra of Nova
Cygni showed the expected strong collisionally excited intercombination lines of ions of
carbon, nitrogen and oxygen, but ratios of lines within the same ion indicated temperatures
(20 000–50 000 K) that were surprisingly high for a nova shell. The same lines were seen in the
planetary nebula spectra and also indicated a high temperature, but in those objects there was
overwhelming evidence that the temperature is actually close to 10 000 K. Mike and co-workers (24) suggested that dielectronic recombination might be exciting some of the lines, thus
distorting the temperature determinations which had assumed that all lines were collisionally
excited, and also affecting the ionization balance. This suggestion was passed on to one of us
(P.J.S.), whose subsequent calculations (Storey 1981) showed that the mechanism of recombination via low-lying autoionizing states, originally discussed in the ionospheric context and
discarded by Bates & Massey (1943), is indeed very important for line formation and the ionization balance in photoionized nebulae such as planetary nebulae, H ii regions and nova shells.
It provided the key to determining the temperature and hence the elemental abundances from
the Nova Cygni 1978 spectra. It is worth noting that the computation of the rate of dielectronic
recombination required a large amount of radiative data for transitions between excited states
of the ions in question, which was readily computed with the general purpose atomic structure
code SUPERSTRUCTURE developed at UCL.
The UCL computer package: some history and personal experiences
of its development
As mentioned above, Mike was instrumental in the acquisition of an IBM 360/65 computer
by UCL in 1965. During the 1960s there was very rapid development in computing technology. One of us (P.J.S.) was taught as an undergraduate how to use a Brunsviga hand calculator, began as a postgraduate under Mike using one of the only two mainframe computers in
London with a one-day turnaround of work and was soon using the powerful IBM 360 located
in the same building. Mike was quick to see the potential of computers in atomic physics and
set in train the process of writing general-purpose programs to compute atomic parameters.
Indeed, Mike’s first task for P.J.S. as a postgraduate in 1966 was to write a computer code to
repeat some calculations of transition probabilities that had appeared in the literature for an
ion of iron. It involved generating angular integrals and varying Slater integrals and spin–orbit
parameters to obtain the optimum match to experimental energies, all of which was ideally
suited to automatic computation. Mike’s thinking became clear when he asked a somewhat
overawed P.J.S. whether it was feasible to generalize the computation to any ion. Within a few
years Werner Eissner, Harry Nussbaumer and Mike Jones, working under Mike’s supervision,
had built a general-purpose FORTRAN atomic structure code, originally called STRUCTURE,
for the computation of energies and radiative transition probabilities. Jones’s contribution was
to add magnetic interactions to the code, essential for the computation of transition probabilities for intercombination transitions, which were being observed in rocket-borne ultraviolet
spectra of the Sun. Probabilities for the forbidden transitions that are the dominant features in
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
13
optical spectra of photoionized nebulae also became relatively easy to calculate. The emphasis
could then move away from the mechanics and labour of making a calculation and shift to
improving the quality of the calculation through configuration interaction, limited only by the
power of the computer available. During the design of SUPERSTRUCTURE, as it was finally
called, there was considerable debate about the choice of one-electron radial wavefunctions
(Nussbaumer & Storey 1983). Mike’s intuition, borne out by experience, was that it would
be more important to pay close attention to configuration interaction rather than the choice of
radial functions, and so Thomas–Fermi-type functions were adopted. SUPERSTRUCTURE
has been, and is still being, used and extended by many people, most notably by Nigel Badnell
in the form of AUTOSTRUCTURE (see, for example, Badnell 1997), which incorporates
the option to compute probabilities for autoionization, making the calculation of dielectronic
recombination processes possible. Badnell was introduced to the problem while a student
under Burgess.
Following the work of Burgess on dielectronic recombination in the solar corona, Mike
became interested in plasma diagnostics with highly ionized species. The forbidden and ultraviolet lines from the corona are excited by electron impact; Mike saw that, for these species, a
great deal of useful work could be done with more approximate methods than close coupling,
and Eissner was set to work writing a general-purpose scattering code using the distortedwave approximation. Helen Mason (Mason 2008) recounts how, when she joined Mike’s
group in 1970 as a postgraduate, she was set the task of using the new distorted-wave code
to compute collision strengths for the six ions of iron and calcium that give rise to prominent
coronal optical forbidden lines. These lines are transitions between the levels of the ground
terms in these ions, but at coronal temperatures direct collisional excitation is augmented by
excitation to higher-lying levels followed by radiative cascade. The new distorted-wave code
made the inclusion of at least some of these higher-lying states possible for the first time.
The parameters calculated by Mason were a marked improvement on what had previously
been available, and the new computer codes permitted a more systematic approach to the
provision of atomic parameters for astrophysics than had previously been possible. At Mike’s
suggestion Mason then turned her attention to the excitation of the ultraviolet and X-ray lines
in the solar spectrum, a new field opening up as the first space-borne spectroscopes began
to return data.
In parallel with the development of the distorted-wave code, Mike was also developing
an approach to the solution of the coupled integro-differential equations of electron–atom
scattering theory referred to earlier. He described an elegant means of converting the integrodifferential equations to linear algebraic equations by using finite differences (19) and later
published the resulting program package, IMPACT (21). The code was used extensively until
the mid 1980s and indeed was used to make the first accurate calculation of collision strengths
for electron excitation of Fe ii, considered a benchmark for difficulty at the time. When Mike
began planning the Opacity Project described in the next section, it became apparent that very
large numbers of photoionization cross-sections would need to be computed, delineating the
energy dependence of the resonance structures that dominate the cross-sections for all except
the simplest atomic ions. With IMPACT, a complete calculation of the radial wavefunctions
was required at each energy, whereas in the R-matrix formulation of the scattering problem
only the asymptotic part of the wavefunction needed to be recalculated. As a consequence
the R-matrix approach was preferred as being more efficient, and from then onwards Mike
concentrated his efforts on developments of that package.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
14
Biographical Memoirs
International Opacity and Iron projects
A major contribution to physics and astronomy made by Mike Seaton was the establishment
of the International Opacity Project, which has rightly been hailed as the ‘crowning achievement of computational atomic physics’. This important undertaking, initiated in 1987 and
described by Mike (27), involved about 30 research workers from five different countries
(France, Germany, the UK, the USA and Venezuela), many of whom were Mike’s previous
research students.
Opacity is a quantity that determines the transport of radiation through matter and is of
crucial importance for many problems in physics and astronomy. Knowledge of the opacity
of stellar material is important in all studies of stellar structure and evolution and of stellar
atmospheres. The opacities are determined by a large number of atomic processes, including
radiative bound–bound, bound–free and free–free transitions.
The most abundant elements in stars are generally hydrogen and helium, for which accurate atomic data are known, but throughout a large part of the envelope regions hydrogen and
helium are fully ionized and the opacity is then largely determined by electrons bound in ions
of heavier elements.
The atomic data required for calculating the opacities caused by these heavier elements
were reviewed by Mike (27), and the computational methods and computer codes used in the
calculations were developed by Mike and collaborators (28). These methods and codes were
based on the close-coupling approximation of electron–atom collision theory using the Rmatrix method, together with new codes for calculating outer-region solutions and dipole integrals. In this approach, configuration space is partitioned into three regions, namely an internal
region, an external region and an asymptotic region, as illustrated in figure 2. In addition, the
target state adopted in this analysis was determined in a separate calculation using either the
code CIV3 (Hibbert 1975) or SUPERSTRUCTURE (Eissner et al. 1974). In the first stage
of the R-matrix calculation in the internal region shown in the figure, the code RMATRXI
(Berrington et al. 1978) was modified by Mike and collaborators (28) to be compatible with
a range of new external and asymptotic region codes developed specifically for the Opacity
Project by Mike. These codes, which calculated solutions correct to second order in the longrange potential in the external and asymptotic regions, were STGF, which calculates collision
strengths and continuum wavefunctions, STGB, which calculates bound-state wavefunctions
and energies, STGD, which calculates damping constants, STGBB, which calculates oscillator
strengths between bound states, STGBF, which calculates bound–free photoionization crosssections, and STGFF, which calculates free–free opacities.
These highly efficient codes were used by Mike and collaborators to calculate the enormous amount of accurate atomic data required in the calculation of opacities. As examples of
these calculations we give in the Bibliography references to the three further papers published
in 1987 (29–31).
Over the next five years the Opacity Project team published more than 25 papers describing the atomic data required for opacity calculations. In addition, two volumes summarizing
the work and achievements of the Opacity Project were published by Institute of Physics
Publishing, in 1995 and 1997. Mike was awarded the Hughes Medal in 1992 by the Royal
Society for this work.
With the results of the Opacity Project in hand, Mike turned his attention to some of the
more subtle effects of stellar opacity. In stellar envelopes, radiation pressure is a significant
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
15
Figure 2. Computer codes used in opacity calculations.
force that acts differently on different ionic species. This leads to diffusive processes that affect
the composition and opacity of the observable atmosphere of the star. The Opacity Project’s
new and accurate Rosseland mean opacities resolved by element and ion were exactly what
was required to solve this problem, and Mike went on to provide the theoretical framework
and computer codes to make this possible (see, for example, (33)).
More recently, as a result of detailed comparisons with work performed by an independent
study of stellar opacities at the Lawrence Livermore Laboratory, called OPAL, the importance
of the contribution from inner-shell transitions to stellar opacities was realized. In a major
new initiative the contribution from these transitions was calculated by Seaton, Badnell and
collaborators (37–39).
Following the success of the International Opacity Project, and taking advantage of the
research workers involved in this project, a major new collaborative research project was proposed and initially led by David Hummer (figure 3). The objectives of this project, reviewed
by Hummer et al. (1993) and called the International Iron Project, was to calculate electronimpact excitation cross-sections and related effective collision strengths for all ionization
stages of iron peak elements, which are of crucial importance in the quantitative analysis of
many astronomical spectra. Of particular importance is singly ionized iron, whose high cosmic abundance, relatively low ionization potential and complex open d-shell atomic structure
ensure that a very large number of electron excited states are observed in objects as diverse
as gaseous nebulae, active galactic nuclei, quasars, Seyfert galaxies and supernova remnants.
Indeed, iron has been called the ‘Rosetta Stone element’ by Lawrence Aller (UCLA), in a
letter to P.G.B. in 1997 emphasizing its importance in so many applications. In conclusion,
this project has been highly successful, resulting so far in more than 70 papers describing the
results, which have been published in the journal Astronomy & Astrophysics.
We note that a review of the International Opacity and Iron projects has been written by
Berrington (1997). In addition, reviews of Mike’s work on stellar atmospheres and modelling, including discussions of opacities, have been written by Butler (2008) and by Delahaye
(2008).
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
16
Biographical Memoirs
Figure 3. Mike and David Hummer relaxing at an Opacity Project meeting in Belfast.
(Photograph provided by Helen Mason.) (Online version in colour.)
Final years in Wales
In 1998, 10 years after Mike formally retired, Mike and Joy moved from their home in London
to Bwlch in the Brecon Beacons National Park. By then Jane was living in the USA, working
as an artist; Richard was in Turkey renovating houses and Tony was in Nuneaton teaching IT
and music.
Mike and Joy’s property has a wonderful view over the Usk valley and includes a holiday
cottage next to their bungalow, where friends and colleagues, including the authors of this
memoir, were always welcome to stay. Joy’s generous hospitality included her delicious speciality of roast Welsh lamb that had been reared locally.
Mike was able to enjoy classical music, particularly Beethoven’s late quartets, reading poetry,
art and novels. He explored the surrounding countryside on walks as long as he could manage.
Mike also continued his research in Wales, keeping in touch with his collaborators at UCL
and all over the world through the Internet. His 80th birthday was celebrated in December
2002 with a symposium at UCL attended by Mike and many of his friends and colleagues. In
2005 he was diagnosed with kidney failure but still managed to continue research until a few
weeks before he died. A group of ex-students and colleagues visited him about a month before
his death, and although his health was failing he presented the group with about 80 pages of
newly written computer code from his latest project. It was clear that he did not expect to be
able to complete it. He died on 29 May 2007 and was buried at the Usk Castle Chase Meadow
natural burial ground in Monmouthshire.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
17
A reminiscence by Joy Seaton
One of the most poignant moments in Michael’s life was during a visit to Kiel in Germany
many years ago. Dining with the physicist Kurt Hunger and his wife one evening, sitting
watching the sunset, the men began talking of their wartime experiences. Kurt had been a captain of a U-boat, Michael in the RAF—the one had sunk many British boats, the other was one
of those who bombed Hamburg. The two men quietly rose from their chairs, clasped their arms
around each other and wept—none of us can ever have forgotten this act of forgiveness.
Final tributes
Finally, we present recollections of Mike as supervisor by two of his former students, P.J.S.
and Claude Zeippen—the latter is now Directeur de Recherches de Première Classe au CNRS
at l’Observatoire de Paris—followed by an extract from a personal tribute made by Claude
Zeippen at an Iron Project meeting in Mons in August 2007, shortly after Mike’s death.
Recalling Mike as a supervisor: Claude Zeippen and Peter Storey
Mike was an awe-inspiring scientist. A meeting with him was always an enriching and sometimes puzzling experience for his students or visitors. Great attention to his explanations was
required as well as patience with his mildly eccentric ways, but the reward was in proportion
to his intellectual powers.
The first time one saw him at length in his office, one understood the words ‘hello’ and
‘cheerio’ and very little in between. It took some concentration to overcome this handicap.
But one soon realized that Mike was available for his students and always willing patiently to
explain easy or difficult points.
An amusing and disconcerting habit that he had was to interrupt his sentences in mid-air
while he took time to think about the rest of his argument, head buried in his hands. One mistake one learned not to make was to try and fill the silences, as Mike would suddenly complete
his sentence regardless of what his visitor had tried to contribute or even if, thinking they were
dismissed, they were about to depart. But not once did one go to him with a problem and come
away without an answer. It could take five minutes or all afternoon; no matter, his patience
was endless and his attention was total.
And he had the kindest possible way of correcting one’s mistakes. On hearing some halfbaked argument he would gently begin ‘I like to think of it like this.…’
The concluding lines from ‘A Tribute by Claude Zeippen’
Here we are and Mike is still with us.…
He will live on as a great moment in the history of atomic physics and atomic astrophysics.
In our work and our lives, he will remain as an example of how much a determined and gifted
individual can accomplish. It is now up to us to preserve and develop his intellectual heritage. He
has taught us not how to dream (Hollywood is so much more efficient in that respect than science…), he has taught us how to try to understand. He has shown us how to be brave and rigorous,
intelligent and astute, motivated and persistent. He will be remembered by many—for he was an
outstanding fellow.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
18
Biographical Memoirs
Honours and awards
1967
Fellow of the Royal Society
1968–72Honorary Editor, J. Phys. B: Atomic and Molecular Physics
1972
Fellow of University College London
1976
Docteur honoris causa from the Observatoire de Paris
1978–81 President of the Royal Astronomical Society
1982Honorary Doctorate from Queen’s University Belfast
1983
Gold Medal from the Royal Astronomical Society
1983Honorary Member of the American Astronomical Society
Guthrie Medal and Prize from the Institute of Physics
1984
Foreign Associate, National Academy of Sciences, USA
1986
1992Hughes Medal, the Royal Society
Acknowledgements
We are grateful to Mike’s wife Joy for information about Mike’s early life and for her memories of the family life
they shared.
The frontispiece photograph was taken by the Royal Astronomical Society when Mike was its president, and is
reproduced with permission.
References to other authors
Aller, L. H., Bowen, I. S. & Minkowski, R. 1955 The spectrum of NGC 7027. Astrophys. J. 122, 62–71.
Badnell, N. R. 1997 On the effects of the two-body non-fine-structure operators of the Breit–Pauli Hamiltonian. J.
Phys. B 30, 1–11.
Baker, J. G. & Menzel, D. H. 1938 Physical processes in gaseous nebulae. III. The Balmer decrement. Astrophys. J.
88, 52–64.
Bates, D. R. 1983 Forbidden atomic lines in auroral spectra. In Atoms in astrophysics (ed. P. G. Burke, W. B. Eissner,
D. G. Hummer & I. C. Percival), pp. 325–345. New York: Plenum Press.
Bates, D. R. & Damgaard, A. 1949 The calculation of the absolute strengths of spectral lines. Phil. Trans. R. Soc.
Lond. A 242, 101–122.
Bates, D. R. & Massey, H. S. W. 1943 The negative ions of atomic and molecular oxygen. Phil. Trans. R. Soc. Lond.
A 239, 269–304.
Berrington, K. A. 1997 The Opacity and Iron Projects—an overview. In Photon and electron collisions with atoms
and molecules (ed. P. G. Burke & C. J. Joachain), pp. 297–312. New York: Plenum Press.
Berrington, K. A., Burke, P. G., LeDourneuf, M., Robb, W. D., Taylor, K. T. & Vo Ky Lan 1978 A new version of
the general program to calculate atomic continuum processes using the R-matrix method. Comput. Phys.
Commun. 14, 367–412.
Brocklehurst, M. 1970 Level populations of hydrogen in gaseous nebulae. Mon. Not. R. Astron. Soc. 148, 417–434.
Brocklehurst, M. 1971 Calculations of the level populations for the low levels of hydrogenic ions in gaseous nebulae.
Mon. Not. R. Astron. Soc. 153, 471–490.
Brocklehurst, M. 1972 The line spectra of helium in gaseous nebulae. Mon. Not. R. Astron. Soc. 157, 211–227.
Burgess, A. 1964 Dielectronic recombination and the temperature of the solar corona. Astrophys. J. 139, 776–780.
Butler, K. 2008 Atomic data for stellar atmospheres. Astron. Geophys. 49, 6.23–6.27.
Delahaye, F. 2008 From atomic data to stellar modelling. Astron. Geophys. 49, 6.28–6.31.
Eissner, W. E., Jones, M. & Nussbaumer, H. 1974 Techniques for the calculation of atomic structures and radiative
data including relativistic corrections. Comput. Phys. Commun. 8, 270–306.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
Michael John Seaton
19
Fano, U. 1970 Quantum defect theory of ℓ uncoupling in H2 as an example of channel-interaction treatment. Phys.
Rev. A 2, 353–365.
Ferland, G. F. 2008 Don Osterbrock and the Seaton–Osterbrock collaboration. Astron. Geophys. 49, 6.32–6.34.
Greene, C. H. & Jungen, C. 1985 Molecular applications of quantum defect theory. Adv. At. Mol. Phys. 21, 51–121.
Hartree, D. R. 1928 The wave mechanics of an atom with a non-Coulomb central field. Part III. Term values and
intensities in series in optical spectra. Proc. Camb. Phil. Soc. 24, 426–437.
Hartree, D. R. 1957 The calculation of atomic structure. New York: John Wiley & Sons Inc.
Hibbert, A. 1975 CIV3—a general program to calculate configuration interaction wave functions and electric-dipole
oscillator strengths. Comput. Phys. Commun. 9, 141–172.
Hummer, D. G., Berrington, K. A., Eissner, W., Pradhan, A. K., Saraph, H. E. & Tully, J. A. 1993 Atomic data from
the Iron Project. 1. Goals and methods. Astron. Astrophys. 279, 298–309.
Jungen, C. & Atabek, O. 1977 Rovibronic interactions in the photoabsorption spectrum of molecular hydrogen and
deuterium: an application of multichannel quantum defect methods. J. Chem. Phys. 66, 5584–5609.
Jungen, C. & Dill, D. 1980 Calculation of rotational–vibrational preionization in H2 by multichannel quantum defect
theory. J. Chem. Phys. 73, 3338–3345.
Mason, H. E. 2008 Emission lines from the solar corona. Astron. Geophys. 49, 620–622.
Moores, D. L. & Saraph, H. E. 1983 Applications of quantum defect theory. In Atoms in astrophysics (ed. P. G. Burke,
W. B. Eissner, D. G. Hummer & I. C. Percival), pp. 173–220. New York: Plenum Press.
Nussbaumer, H. & Storey, P. J. 1983 The University College computer package. In Atoms in astrophysics (ed. P. G.
Burke, W. B. Eissner, D. G. Hummer & I. C. Percival), pp. 265–288. New York: Plenum Press.
Storey, P. J. 1981 Dielectronic recombination at nebular temperatures. Mon. Not. R. Astron. Soc. 195, 27P–31P.
Bibliography
The following publications are those referred to directly in the text. A full bibliography is
available as electronic supplementary material at http://dx.doi.org/10.1098/rsbm.2012.0044
or via http://rsbm.royalsocietypublishing.org.
(1)
1953The Hartree–Fock equations for continuum states with applications to electron excitation of the ground
configuration terms of O i. Phil. Trans. R. Soc. Lond. A 245, 469–499.
(2)Electron excitation of forbidden lines occurring in gaseous nebulae. Proc. R. Soc. Lond. A 218,
400–416.
(3) 1955Le calcul approximatif des sections efficaces de photo-ionisation atomiques. I. L’emploi des functions
d’onde hydrogenoides pour les états continus. C. R. Acad. Sci. Paris 240, 1193–1195.
(4) 1956The calculation of cross-sections for excitation of forbidden atomic lines by electron impact. In The
airglow and aurora (J. Atmos. Terr. Phys. suppl. 5) (ed. E. B. Armstrong & A. Dalgarno), pp. 289–301.
London: Pergamon Press.
(5) 1957 (With D. E. Osterbrock) Relative [O ii] intensities in gaseous nebulae. Astrophys. J. 125, 66–83.
(6)
(With I. C. Percival) The partial wave theory of electron hydrogen atom collisions. Proc. Camb. Phil.
Soc. 53, 654–662.
(7) 1958The quantum defect method. Mon. Not. R. Astron. Soc. 118, 504–518.
(8) 1960 (With A. Burgess) A general formula for the calculation of atomic photo-ionization cross-sections.
Mon. Not. R. Astron. Soc. 120, 121–151.
(9)
(With L. Castillejo & I. C. Percival) On the theory of elastic collisions between electrons and hydrogen
atoms. Proc. R. Soc. Lond. A 254, 259–272.
(10)
Planetary nebulae. Rep. Prog. Phys. 23, 313–354.
(11)H i, He i and He ii intensities in planetary nebulae. Mon. Not. R. Astron. Soc. 120, 326–337.
(12) 1964 (With R. J. Harman) The central stars of planetary nebulae. Astrophys. J. 140, 824–828.
(13)
(With R. M. Pengelly) Recombination spectra. II. Collisional transitions between states of degenerate
energy levels. Mon. Not. R. Astron. Soc. 127, 165–175.
Downloaded from http://rsbm.royalsocietypublishing.org/ on June 17, 2017
20
Biographical Memoirs
Recombination spectra. III. Populations of highly excited states. Mon. Not. R. Astron. Soc. 127,
177–184.
(15) 1966 Quantum defect theory. I. General formulation. Proc. Phys. Soc. 88, 801–814.
(16) 1969 Quantum defect theory. VII. Analysis of resonance structures. J. Phys. B 2, 5–11.
(17) 1971 (With P. G. Burke) Numerical solutions of the integro-differential equations of electron–atom collision theory. In Methods in computational physics, vol. 10 (ed. B. Alder, S. Fernbach & M. Rotenberg),
pp. 1–80. New York: Academic Press.
(18) 1972 (With M. Brocklehurst) On the interpretation of radio recombination line observations. Mon. Not. R.
Astron. Soc. 157, 179–210.
(19) 1974Computer programs for the calculation of electron–atom collision cross sections. II. A numerical
method for solving the coupled integro-differential equations. J. Phys. B 7, 1817–1840.
(20) 1976 (With P. J. Storey) Di-electronic recombination. In Atomic processes and applications (ed. P. G. Burke
& B. L. Moiseiwitsch), pp. 133–197. Amsterdam: North-Holland.
(21) 1978 (With M. A. Crees & P. M. H. Wilson) IMPACT, a program for the solution of the coupled integro-differential equations of electron–atom collision theory. Comput. Phys. Commun. 15, 23–83.
(22)
Quantum defect theory. XI. Clarification of some aspects of the theory. J. Phys. B 11, 4067–4093.
(23) 1979Interstellar extinction in the UV. Mon. Not. R. Astron. Soc. 187, 73P–76P.
(24) 1980 (With J. P. Harrington, J. H. Lutz & D. J. Stickland) Ultraviolet spectra of planetary nebulae. I. The
abundance of carbon in IC 418. Mon. Not. R. Astron. Soc. 191, 13–22.
(25) 1983 Quantum defect theory. Rep. Prog. Phys. 46, 167–257.
(26) 1985 (With R. H. Bell) Dielectronic recombination. I. General theory. J. Phys. B 18, 1589–1629.
(27) 1987Atomic data for opacity calculations. I. General description. J. Phys. B 20, 6363–6378.
(28)
(With K. A. Berrington, P. G. Burke, K. Butler, P. J. Storey, K. T. Taylor & Yu Yan) Atomic data for
opacity calculations. II. Computational methods. J. Phys. B 20, 6379–6397.
(29)
(With Yu Yan & K. T. Taylor) Atomic data for opacity calculations. III. Oscillator strengths for C ii. J.
Phys. B 20, 6399–6408.
(30)
(With Yu Yan) Atomic data for opacity calculations. IV. Photoionization cross sections for C ii. J. Phys.
B 20, 6409–6429.
(31)Atomic data for opacity calculations. V. Electron impact broadening of some C iii lines. J. Phys. B 20,
6431–6446.
(32) 1996Levitation. Astrophys. Space Sci. 237, 107–123.
(33) 1997 Radiative accelerations in stellar envelopes. Mon. Not. R. Astron. Soc. 289, 700–720.
(34) 2002Coulomb functions for attractive and repulsive potentials and for positive and negative energies.
Comput. Phys. Commun. 146, 225–249.
(35)
FGH, a code for the calculation of Coulomb radial wavefunctions from series expansions. Comput.
Phys. Commun. 146, 250–253.
(36)NUMER, a code for Numerov integrations of Coulomb functions. Comput. Phys. Commun. 146,
254–260.
(37) 2003 (With N. R. Badnell) On the importance of inner-shell transitions for opacity calculations. J. Phys.
B 36, 4367–4385.
(38) 2004 (With N. R. Badnell) A comparison of Rosseland-mean opacities from OP and OPAL. Mon. Not. R.
Astron. Soc. 354, 457–465.
(39) 2005 (With N. R. Badnell, M. A. Bautista, K. Butler, F. Delahaye, C. Mendoza, P. Palmeri & C. J. Zeippen)
Updated opacities from the Opacity Project. Mon. Not. R. Astron. Soc. 360, 458–464.
(14)