www.elsevier.com/locate/asr

Advances in Space Research 36 (2005) 1415–1421

EUV spectroscopy for Solar Orbiter

R.A. Harrison

Space Science and Technology Department, Rutherford Appleton Laboratory, Chilton, Didcot, Oxfordshire OX11 0QX, UK

Received 28 July 2004; received in revised form 30 November 2004; accepted 10 December 2004

Abstract

Building on the success of EUV/UV spectroscopic studies fromx the SOHO mission in particular, a next generation spectroscopic

study of the Sun is included in plans for Solar Orbiter. The combination of close-up and out of ecliptic observation provides unique

possibilities for solar plasma diagnostics and these are outlined here. Technical challenges and the instrumental requirements for

such an instrument aboard Orbiter are described in detail.

� 2004 COSPAR. Published by Elsevier Ltd. All rights reserved.

Keywords: Sun; Solar activity; Solar atmosphere; EUV spectroscopy

1. Introduction

Spectroscopic observations of emission lines in the

EUV/UV region of the electromagnetic spectrum are

critical for the determination of plasma diagnostics from

the solar atmosphere, providing the necessary tools for

probing the wide solar plasma temperature range, fromtens of thousands to several million K. Analysis of the

emission lines, mainly from trace elements in the Sun�satmosphere, provides information on plasma density,

temperature, element/ion abundances, flow speeds and

the structure and evolution of atmospheric phenomena.

Such information provides a foundation for understand-

ing the physics behind a large range of solar phenomena.

Therefore, it is logical that such instrumentation shouldbe included in the Solar Orbiter strawman payload.

Current spacecraft instrumentation (SOHO, see Dom-

ingo et al., 1995, and references therein) provides EUV

spatial and spectral resolving elements of order 2 arc-

sec/pixel and 0.01 nm/pixel, respectively, and UV resolv-

ing elements of 1 arcsec/pixel and 0.002 nm/pixel. There is

no EUV or UV spectroscopic capability aboard the

NASA STEREOmission (2006 launch), the NASA Solar

0273-1177/$30 � 2004 COSPAR. Published by Elsevier Ltd. All rights reser

doi:10.1016/j.asr.2004.12.025

E-mail address: r.harrison@rl.ac.uk.

Dynamics Observatory (2008 launch) and the NASA

Solar Probe (currently under study). The only planned

EUV spectrometer for an approved future solar mission

at this time is the EIS instrument on Solar-B with 1 arc-

sec/pixel (750 km on the Sun) performance (Shimizu,

2002). We note that the wavelength selection of EIS is

tuned to higher temperature, coronal, plasmas, particu-larly well suited to studies of active regions and even

flares, rather than the quiet solar atmosphere in general.

Figs. 1 and 2 show a brief glimpse of the power of

EUV spectroscopy in studying the solar atmosphere.

Fig. 1 shows the Sun imaged in emission from six differ-

ent emission lines, effectively revealing six temperature

regimes at the same time, in the range 20,000 to

2,000,000 K. Fig. 2 shows a close up of an active regionon the western limb. A bright loop is clearly visible at

transition region temperatures. Analysis of the emission

line shifts show that above the loop there is a radial jet

of plasma which is spiralling outward from the Sun. The

ability to select particular emission lines for specific

analyses, to reveal temperatures, or densities, and spec-

tral analyses designed to reveal flow properties, are par-

ticularly valuable for determining the plasma propertiesof the solar atmosphere. The application of spectros-

copy in the EUV to solar phenomena is well demon-

ved.

Fig. 1. SOHO CDS images of the Sun taken in six emission lines from

helium, oxygen, magnesium and iron, representing temperatures from

20,000 K (top left) to 2,000,000 K (bottom right).

Fig. 2. A close-up of an active region loop, taken in the 250,000 K

emission line at 629 A from O V (oxygen ionised 4 times). Whilst the

left image shows the bright loop, a Doppler analysis of the emission

line reveals a spiral jet of plasma streaming vertically away from the

region.

1416 R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421

strated by the publication of many hundreds of papersto date using the CDS and SUMER instruments on

SOHO (see sohowww.nascom.nasa.gov).

Solar Orbiter provides a unique platform for solar

spectroscopy, with close encounters and high latitude

Table 1

Instrument requirements and allocations

Goal

Spatial resolving element 1.0 arcsec pixels, or better

Spectral resolving element 0.02 A/pixel or better

Field of view 20 arcmin · 20 arcmin or larger

Wavelength prime bands 170–220, 580–630, >912 A

Pointing To anywhere on disc and low corona off-disc

Mass 25 kg

Telemetry 17 kbit/s

Autonomy Pre-planned sequences in deferred command sto

Stabilisation system May require active on-board system; on-ground

capabilities. SOHO has demonstrated well the power

of a combination of imaging and spectroscopy and this

has been a basic driver for the remote sensing package

on Orbiter. However, such a mission does pose signifi-

cant problems for instrument design, development and

operation, with particular issues including thermal ex-tremes, severe particle environments, a requirement for

autonomy during solar passes, and limitations on mass

and telemetry. An overview of the Solar Orbiter mission

is given by Marsch et al. (2001).

2. Instrument requirements

At this stage, the requirements for an EUV or UV

spectrometer on Solar orbiter are somewhat flexible.

However, Table 1 lists the major goals and limitations

as we understand them at present, and the allocations

for telemetry, mass, etc. Many of the items are discussed

in greater detail below.

3. Instrument optical design concept

Whilst recognising that an EUV/UV spectrometer is

an essential component of the Solar Orbiter, we must

be aware that the mission concept demands that it must

be compact and light-weight, must not be too telemetry

�thirsty� and must be able to cope with the thermal and

particle environment of such an orbit.A normal incidence system was originally envisaged

for this spectrometer, to fit within reasonable length lim-

its of the spacecraft. However, the thermal situation is

extreme, in particular for a normal incidence design.

The basic design we propose here is an off-axis normal

incidence (NIS) system where a single paraboloid pri-

mary mirror reflects the selected portion of the solar im-

age through a heat-stop to a spectrometer using avariable line spaced (VLS) grating in normal incidence.

The concept is shown in Fig. 3 The wavelength selec-

tions are geared to bright solar lines in the EUV from

a broad range of temperatures. The basic design concept

is presented by Thomas (2003).

Comment

150 km on the Sun at 0.2 AU

To separate emission lines and for flow studies

Size of active region from 0.2 AU

Prime bands to cover emission lines of interest

Co-pointed with remote sensing package

Due to spacecraft limitations

Current allocation

re No contact during solar encounters

option under study Spacecraft performance needs to be defined

Fig. 3. The basic design concept of the EUV spectrometer.

R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421 1417

In this concept, the primary optical component is a

paraboloid mirror. The off-axis approach allows the

insertion of a heat-stop between the primary and the slit,

which only allows a selected area of the solar image into

the spectrometer. Most of the solar radiation is reflected

by the heat-stop out of the front aperture. Thus, thethermal challenge for this design is almost exclusively

concerned with the thermal control of the primary mir-

ror itself. We address this later.

The slit assembly lies at the focal plane, below the

heat-stop, and beyond this is the spectrometer, with a

toroidal VLS grating, forming a focus at a 2-D detector.

There is no secondary mirror, as with a Ritchey–Chret-

ien design, for example, and this helps to maintain a rea-sonable effective area. The VLS grating approach allows

good off-axis performance compared to a uniform grat-

ing, and it brings the spectrometer �arm� closer to the

axis of the instrument, making the envelope smaller.

The grating ruling spacing is yet to be decided but values

up to 4800 l/mm have been considered for design inves-

tigations. The use of a toroidal VLS grating allows a

spectrometer magnification, with values of 2.5 beingconsidered, whilst retaining optical quality.

Several wavelength bands are under consideration at

this time and have been discussed widely (e.g. Harrison

and Vial, 2001). Prime regions of the EUV spectrum

have been identified, which contain emission lines of

particular importance, and these include 170–220, 580–

630 and >912 A. These three would allow detailed stud-

ies of coronal, transition region and chromospheric plas-mas. Obtaining spectra from three such bands would be

very difficult, so we anticipate obtaining two of these

bands using either two orders or two detectors at the fo-

cus of the spectrometer.

The primary mirror presents a portion of the Sun at

the slit, and it is this mirror that can be rotated to allow

rastered images, i.e., exposures interlaced with mecha-

nism movements to build up images simultaneously inselected wavelengths. Only a small fraction of the Sun

will pass through the heat-stop to the slit assembly; pos-

sibly of order several hundredths of the disc area.

The instrument will not have independent pointing. It

will be hard-mounted with the other remote sensing

instruments, and used in conjunction with those instru-

ments. Limited independent pointing can be performedusing the primary mirror mechanism.

The instrument would include a selection of slits,

which can be chosen for particular observation pro-

grammes. In addition, it is noted that the instrument res-

olution may demand an image stabilisation system

which will involve motion of the primary mirror in re-

sponse to an external signal.

We note that what is presented here is one possibleoptical concept, which is very much based on heritage

from the SOHO-CDS instrument. At this stage, the

instruments are not yet selected. In addition, we point

out that whilst the thermal aspects of the normal inci-

dence design are being investigated, a second design op-

tion based on a grazing incidence Wolter II design is

also being considered. This would also include a variable

line spaced grating and similar detectors, but the ther-mal extremes encountered by such a design would be less

severe.

4. Resolution and the detector system

We size the instrument to an observing distance of

0.2 AU (perihelion). A spatial pixel size of 1 arcsec rep-resents 750 km on the Sun from the Earth; the same pix-

el at 0.2 AU represents 150 km. This is an order of

magnitude better than the CDS instrument pixel size

on SOHO (Harrison et al., 1995), a factor of five better

than the SUMER instrument on SOHO (Wilhelm et al.,

1995) and about half the size of the best EUV imaging

capability (TRACE, 0.5 arcsec pixels at 1 AU). Our ba-

sic aim is to achieve 1 arcsec pixel elements, recognisingthat (a) the limitations of mass and telemetry demand a

compact instrument with little complexity, in terms of

the basic optical layout, operation, and the use of mech-

anisms, etc. and (b) the novelty of the mission is really

derived from the orbital path rather than major instru-

ment advances. We must be aware that Orbiter is not

a mission that can carry the large, complex instruments

one might expect from a near-Earth mission.Initial discussions included provision for a 0.5 arcsec

pixel (75 km on the Sun). This led to the consideration

of a detector array baseline of 4k · 4k 5 lm pixels. This

would also suggest that the instrument would have a

spectral range of 4 nm at 0.001 nm/pixel. The same array

will give a spatial extent (vertical distance on the detec-

tor = slit length) of 34 arcmin. The solar diameter at

0.2 AU is 170 arcmin, i.e., we have a slit length of 0.2of the solar diameter. For a given pointing location

(spacecraft pointing), rastered imaging will be made up

1418 R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421

from movement in one direction of the primary mirror.

Thus, the basic rastered �field of view� would be 34 · 34

arcmin. This would be a size somewhat larger than a

large active region, i.e., somewhat larger than the CDS

field of view on SOHO.

The demand for 0.5 arcsec pixels (5 lm) has been re-laxed, with recent studies suggesting that a target of 1.0

arcsec (150 km on the Sun) is appropriate, i.e., it leads to

an instrument option whose size and mass is more con-

sistent with the spacecraft limitations. Although further

study is required, this may result in a 2k · 2k detector of

pixel size 10 lm. A parallel discussion about the field of

view has stressed that the minimum size should be 20

arcmin · 20 arcmin.The choice of detector is dictated by the harsh parti-

cle environment which will be encountered by Solar Or-

biter, as well as mass and power constraints. The

particle environment, which will be encountered by Or-

biter, means that CCD-type detectors will most likely be

inappropriate. We may anticipate a solar wind �back-ground� proton flux some 25· that of SOHO (1/r2).

For an average flux at 1 AU, of density 9 cm�3 (averagespeed and temperature of 300 km/s and 4 · 105 K

(3.5 keV)) we expect 225 cm�3 at 0.2 AU. Thus the nom-

inal particle environment will be similar to some modest

storm events detected occasionally by SOHO.

There may also be an increased chance of encounter-

ing proton �storms�, due to vicinity, from shocks associ-

ated with mass ejection, with up to thousands of proton

hits per second. One might expect events similar to thoseexperienced by SOHO, with greater intensity, and, in

addition, some near-Sun events may be generated by lat-

eral expansion of CME disturbances. The exact intensi-

ties remain unknown. The geometrical factors and

magnetic configurations, which may play a role in defin-

ing the chance of occurrence of storms are also ill

defined.

Also, we anticipate occasional impacts from solarflare neutrons whose 15.5 min lifetime means that most

missions do not encounter them. Finally, we anticipate a

similar cosmic ray (non solar) flux similar to that at

SOHO.

The net effect is an increase in particle hits, with some

extreme conditions including occasional neutrons. The

radiation damage in CCDs is mostly caused by the cre-

ation of charge traps reducing the charge transfer effi-ciency (CTE). The radiation hardness of silicon Active

Pixel Sensor (APS) detectors is much higher because

CTE degradation is unimportant; charge is not trans-

ferred across the array using an APS detector, where

on-chip electronics allows the extraction and amplifica-

tion of charge from each pixel individually. The charge

collection efficiency (CCE) may also degrade, but at

higher radiation levels.The APS detector system is a realistic option for So-

lar Orbiter from a particle environment point of view.

We note, however, that the on-chip electronics also pro-

vides additional low mass and power advantages, com-

pared to CCDs. The APS EUV sensitivity will be

provided in the same way as with CCDs, in this case

with back-thinned devices. As with CCDs, the APS sys-

tem would need to be cooled to about �80�, using a ded-icated radiator.

The development of the APS system, specifically for

Solar Orbiter applications is discussed by Prydderch et

al. (2004).

5. Thermal issues

At 1 AU the average solar intensity is 1.371 kW/m2.

During the nominal phase of the Solar Orbiter mission,

the spacecraft occupies a 149 day orbit with aphelion

and perihelion 0.8 and 0.2 AU, respectively. Thus, every

75 days, the spacecraft will encounter a range from

2.142 kW/m2 (0.8 AU) to 34.275 kW/m2 (0.2 AU – 25

times the value at 1 AU). This presents a severe thermal

challenge, which we tackle in a number of ways.The primary mirror ‘‘sees’’ the full Sun. It is a com-

monly held view that such a mirror would only view

the instrument field of view, but the opening angle is

such that the full Sun is visible to the mirror. The limits

on instrument length do not allow provision for a colli-

mated aperture. One option being studied is the use of

an uncoated SiC mirror. SiC optical components can

run hotter than traditional components. The originalmirror size under study was a 120 mm aperture, result-

ing in the primary receiving about 390 W at 0.2 AU.

Coatings can be used to reduce the absorption (to 0.2

for gold coating), though another option is to make

the primary absorbing to reduce the thermal effects later

in the optical system. Various options are under consid-

eration. However, running the mirror at 61 �C, with a

fixed radiator temperature of 50 �C, with a SiC mirror(absoptivity 0.1), the total radiator area required is

0.19 m2, i.e., 0.43 m · 0.43 m. However, this is a preli-

minary estimate, used as an example, and it does not in-

clude the radiator area for the APS detector. It does not

include, also, any consideration of the thermal interface

to the spacecraft. If the radiator sees radiation from the

back of the spacecraft heat-shield, for example, the radi-

ator size must increase. However, this estimate does sug-gest that the radiator size will be feasible, that is, within

the footprint of the instrument.

These estimates are for the situation at 0.2 AU. How-

ever, the thermal input varies considerably over the 149

day orbit – by a factor of 16. This can produce large

temperature variations over the orbit. To cope with this,

a combination of heat-switches, heaters and heat-pipes

are being considered. However, we note that, accordingto the Solar Orbiter mission concept, the instrument will

not run its prime scientific operation outside the solar

R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421 1419

encounter (30 day) periods, so it is not necessary to

strive to attain perfect optical alignment during the aph-

elion periods. Some flexing of the instrument is antici-

pated, with the best optical performance geared to the

solar encounter periods.

6. Other issues

The total allocated mass, at 25 kg, is light compared

to similar instruments in operation. For example, the

SOHO CDS instrument has a mass of 100 kg. However,

CDS is a double spectrometer with an aluminium alloy

structure, 5 individual detector systems and an indepen-dent pointing system. The spectrometer under discus-

sion here will be a single spectrometer, with a smaller

structure, made of carbon fibre, with SiC optics, one

detector system and no independent pointing system.

The Solar-B EIS instrument (Culhane et al., 2002) under

development for the Solar-B mission weighs in at 60 kg,

but is 3 m long. Thus, a mass under 30 kg for a modern

instrument 1.0–1.5 m in length appears to be a chal-lenge, but within the realms of possibility.

Preliminary mass calculations have indicated masses

of 25–30 kg but some refinement of the design and opti-

misation of the structural concept have yet to be com-

pleted. Basic decisions such as the inclusion of an

image stabilisation system need to be made.

The nominal telemetry rate for the Orbiter spectrom-

eter is 17 kbit/s. If we assume a detector image of4k · 4k pixels, at 12 bits per pixel, it will take 197 min

to read one exposure. Since each exposure will form part

of a rastered image, the raster cadence will be signifi-

cantly longer. These sobering figures stress the extreme

problem facing such instruments. The situation is better

for the 2k · 2k array, but it is still a major problem.

However, this is not a new problem.

Studies from instruments such as CDS have shownthat careful line selection is far more important than

data compression in managing the data return of such

a spectrometer. Much of the spectrum is not required.

Indeed, for specific studies, specific emission lines are re-

quired. A good rule of thumb from SOHO is that a

selection of between 6 and 15 lines is good for most sci-

entific purposes.

Let us assume a resolving element of 1.0 arcsec alonga 2k pixel slit and a nominal spectral resolution of order

0.02 A/pixel. To obtain full line widths for million K

lines, plus sufficient nearby background, one would

want to return about 0.3 A, i.e., 15 pixels. Thus, to re-

turn data for 6 emission lines along the full length of

the slit would take 127 s. Assuming a factor of 3 com-

pression, and a rastered image of 50 mirror locations,

and taking 50 pixels along the slit, to produce a final ras-ter sequence of 50 arcsec · 50 arcsec, the raster cadence

would become 2.6 min.

Such basic calculations show that with careful selec-

tion and compression, and some flexibility in the meth-

ods for selection and compression, we can achieve

reasonable data return rates. It does stress the need for

careful planning in terms of line selection and target

selection. Much depends on the particular study in ques-tion; some applications require long exposures and slow

cadences and some require rapid data return. In general,

for the latter we must use small area rasters and few

lines.

However, for many studies, we do need to consider

methods for improving the performance to avoid com-

promising the scientific return. Several options are pos-

sible, including:

� Increasing the telemetry rate: This option should be

sought from the Project in any case because the

telemetry rates are calculated on assumptions on

on-board memory (240 Gbit) and ground station sup-

port (one ground station) which were made two years

ago.

� Returning line profile parameters rather than the fullprofile information.

� Returning image differences rather than full images.

These are just three options, the first one of which is

beyond the control of the instrument study team!

Given our plan to achieve 1.0 arcsec resolution ele-

ments, we must consider the inclusion of an image sta-

bilisation system. The need for this will be dictated bythe final spacecraft stability performance. We anticipate

a stability performance which will make the inclusion of

such a system a marginal requirement. An on-board sys-

tem has been discussed with the assumption that image

pointing information be made available from the high

resolution VIM instrument and the primary mirror

would be tilted in response. This is akin to the image sta-

bilisation system used on the TRACE spacecraft. An-other option under careful consideration is to not

include an image stabilisation system at all, assuming

that the variations of the spacecraft stability occur on

timescales much less than the exposure time of the spec-

trometer and thus any corrections could be done on the

ground. Such an after-the-event correction saves mass

and has been used in small-scale event analyses quite

successfully for the SOHO CDS instrument.For the present, the latter is assumed, but both op-

tions must remain open.

EUV optical surfaces in space are renowned for expe-

riencing significant degradation in performance with

time. Under high irradiation, particularly in the ultravio-

let, any contaminant deposited on an optical surface, even

in veryminute amounts, polymerizes, so the reflectivity of

the surface drastically decreases. This effect is well knownfor synchrotron radiation optics as well as for some space

instruments, but has been well avoided by the SOHO

1420 R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421

EUV/UV instruments. The degree to which the reflectiv-

ity decreases depends on the irradiation exposure and

on the partial pressure of the contaminant.

However, for Solar Orbiter the situation is more dif-

ficult than it was for SOHO; due to the changing dis-

tance from the Sun, the level of UV irradiation will behigher and the thermal environment more variable.

Even with the most stringent procedures in the han-

dling and assembling of the optical components, under

the extreme irradiation conditions at 0.2 AU, there is a

risk of a serious rapid degradation of the reflectivity, espe-

cially in the EUV. The variable thermal environment dur-

ing the orbit makes evaporation and out-gassing from

surfaces with increasing temperature unavoidable.The decreasing reflectivity could be severe for optics at

normal incidence, where the EUV reflectivity is relatively

low and the EUV absorption is high. For example, gold

could be a good candidate as an EUV coating for mirrors

at normal incidence, since it has high visible reflectivity

and also discrete EUV reflectivity, but a thin layer of con-

taminants deposited on its surface could drastically re-

duce the EUV response, and thus the effective area.It should be noted that the mirrors can be operated at

relatively high temperature and this could help to reduce

the deposition of contaminants.

It should be noted, also, that steps can be taken to re-

duce the levels of potential contamination, in space. The

most important procedure would be a long out-gassing

period prior to opening the instrument door. For the

CDS and SUMER instruments on SOHO, the out-gas-sing period was 3 months from launch, and this was a

deliberate (and successful) policy. With the inclusion

of vents allowing out-gassing materials to escape, the

long period certainly enabled the contamination to be

reduced. Such a policy must be adopted for Solar Orbi-

ter – possibly for several instruments. For efficient vent-

ing, the opening to space must be large (e.g., a partly

opened aperture door, a door specifically designed forventing, or a permanent vent) and, in addition, the

instrument interior must be preferentially heated (by

passive or active heating).

Any EUV instrumentation must be developed with

the most stringent contamination policy, both in the lab-

oratory and in operation (e.g., the out-gassing men-

tioned above). Possible effects must be assessed

thoroughly by the proposing teams and optical and pro-cedural policies adopted.

7. Operation

The spectrometer instrument requires to be Sun

pointed. It will be hard-mounted, along with the other

remote sensing instruments. The required co-alignmentaccuracy between instruments is 2 arcmin, based on

attaining a reasonable image overlap with the smallest

instrument field of view. In addition, in operation, a

pointing accuracy of 2 arcmin is required. Fine pointing

within the field of view of the instrument can be

achieved using the mirror mechanism.

Operations will be performed in pre-planned se-

quences stored, time-tagged in a deferred commandstore. The operations for an entire encounter (30 days)

should be stored in this way. The sequences will have

been selected during the period preceding the solar

encounter. The planning and the selection of sequences

will be done in concert with the other remote-sensing

instruments. Indeed, target selection and pointing will

be done as one instrument.

During the non-solar encounter periods of the orbit(the period outside the 30 days observation) the stored

observational data will be telemetered to the ground.

However, it would be highly desirable, if not essential,

to practice encounter observations – even with consider-

ably reduced telemetry – much as planetary encounter

missions do, during the non-encounter parts of the or-

bit. In addition, any telemetry allocation, however

low, for some modest synoptic observations during therest of the orbit would be desirable.

8. Final remarks

This report is necessarily both preliminary and vague.

Many decisions need to be taken before the spacecraft

and mission concept is refined, and this influences theinstrument design. Some areas of instrument develop-

ment require technical development, the outcome of

which will be exploited to the full, and we can only

anticipate the outcome of study work at this time. Even

the scientific requirements are under discussion at this

stage. However, Solar orbiter is an approved mission

and the aim is to lay the foundation for the inclusion

of an EUV spectrometer which builds upon our experi-ence gained with the SOHO mission. The instrument de-

scribed here is both innovative (for example, in the use

of APS detectors and VLS gratings) and sufficiently ro-

bust and basic, as required for a mission with such an

array of technical challenges.

References

Culhane, J.L., Doschek, G.A., Watanabe, T., Lang, J., The EUV

imaging spectrometer and its role in the Solar-B mission, in:

Pauluhn, A., Huber, M.C.E., von Steiger, R. (Eds.), The Radio-

metric Calibration of SOHO. ISSI Scientific Report, SR-002, ISSN

1608-280X.

Domingo, V., Fleck, B., Poland, A.I. The SOHO mission: an overview.

Solar Phys. 162, 1, 1995.

Harrison, R.A., Vial, J.-C., Solar Orbiter EUV/UV wavelength

selection and instrumentation, in: Proceedings of the Solar

Encounter: The First Solar Orbiter workshop, ESA SP-493, p.

151, 2001.

R.A. Harrison / Advances in Space Research 36 (2005) 1415–1421 1421

Harrison, R.A., Sawyer, E.C., Carter, M.K., et al. The Coronal

Diagnostic Spectrometer for the solar and heliospheric observa-

tory. Solar Phys. 162, 233, 1995.

Marsch, E., Antonucci, E., Bochsler, P., Bougeret, J.-L., Fleck, B.,

Harrison, R.A., Marsden, R., Schwenn, R., Vial, J.-C. Solar

orbiter, a high-resolution mission to the Sun and inner heliosphere,

in: Recent insights into the physics of the Sun and heliosphere, IAU

Symposium, vol. 203, p. 565, 2001.

Prydderch, M., Waltham, N.R., Morrissey, Q., Turchetta, R., Pool, P.

A large area CMOS monolithic active pixel sensor for extreme UV

spectroscopy and imaging, Proc. SPIE 5301, 175, 2004.

Shimizu, T. Solar-B. Adv. Space Res. 29 (12), 2009, 2002.

Thomas, R.J. Toroidal varied line-space (TVLS) gratings, Proc. SPIE

4853, 411, 2003.

Wilhelm, K., Curdt, W., Marsch SUMER – Solar Ultraviolet

Measurements of Emitted Radiation. Solar Phys. 162, 189, 1995.