Daniel Wood

In situ trap parameter studies in CCDs for space applications

Charge-Coupled Devices are the detector of choice for the focal planes of many optical and X-ray space telescopes. In recent years, EM-CCDs, SCDs and CMOS sensors have been used, or baselined, for missions in which the detection of X-ray and visible photons are key to the science goals of the mission. When placed in orbit, silicon-based detectors will suffer radiation damage as a consequence of the harsh space radiation environment, creating traps in the silicon. The radiation-induced traps will capture and release signal electrons, effectively “smearing” the image. Without correction, this smearing of the image would have major consequences on the science goals of the missions. Fitting to observed results, through careful planning of observation strategies while the radiation dose received remains low in the early stages of the mission, has previously been used to correct against the radiation damage effects. As the science goals becoming increasingly demanding, however, the correction algorithms require greater accuracy and a more physical approach is required, removing the effects of the radiation damage by modelling the trap capture and release mechanisms to a high level of detail. The drive for increasingly accurate trap parameters has led to the development of new methods of characterisation of traps in the silicon, measuring the trap properties and their effects to the single-trap level in situ. Here, we summarise the latest developments in trap characterisation techniques for n-channel and p-channel devices.

A study of the double-acceptor level of the silicon divacancy in a proton irradiated n-channel CCD

Radiation damage effects are problematic for space-based detectors. Highly energetic particles, predominantly from the sun can damage a detector and reduce its operational lifetime. For an image sensor such as a Charge-Coupled Device (CCD) impinging particles can potentially displace silicon atoms from the CCD lattice, creating defects which can trap signal charge and degrade an image through smearing. This paper presents a study of one energy level of the silicon divacancy defect using the technique of single trap-pumping on a proton irradiated n-channel CCD. The technique allows for the study of individual defects at a sub-pixel level, providing highly accurate data on defect parameters. Of particular importance when concerned with CCD performance is the emission time-constant of a defect level, which is the time-scale for which it can trap a signal charge. The trap-pumping technique is a direct probe of individual defect emission time-constants in a CCD, allowing for them to be studied with greater precision than possible with other defect analysis techniques such as deep-level transient spectroscopy on representative materials.

In situ trap properties in CCDs: the donor level of the silicon divacancy

The silicon divacancy is one of the main defects of concern in radiation damage studies of Charge-Coupled Devices (CCDs) and, being immobile at room temperature, the defect is accessible to a variety of characterisation techniques. As such, there is a large amount of (often conflicting) information in the literature regarding this defect. Here we study the donor level of the divacancy, one of three energy levels which lie between the silicon valence and conduction bands. The donor level of the divacancy acts as a trap for holes in silicon and therefore can be studied through the use of a p-channel CCD. The method of trap-pumping, linked closely to the process of pocket-pumping, has been demonstrated in the literature over the last two years to allow for in-situ analysis of defects in the silicon of CCDs. However, most work so far has been a demonstartion of the techinique. We begin here to use the technique for detailed studies of a specific defect centre in silicon, the donor level of the divacancy. The trap density post-irradiation can be found, and each instance of the trap identified independently of all others. Through the study of the trap response at different clocking frequencies one can measure directly the defect emission time constant, and through tracking this at different temperatures, it is possible to use Shockley-Read-Hall theory to calculate the trap energy level and cross-section. A large population of traps, all with parameters consistent with the donor level of the divacancy, has been studied, leading to a measure of the distribution of properties. The emission time constant, energy level and cross-section are found to have relatively large spreads, significantly beyond the small uncertainty in the measurement technique. This spread has major implications on the correction of charge transfer inefficiency effects in space applications in which high precision is required.

Mapping radiation-induced defects in CCDs through space and time

The Charge Coupled Device (CCD) has long been the detector of choice for many space-based applications. The CCD converts the signal X-rays or visible light into electrons (n-channel devices) or holes (p-channel devices) which are stored in the pixel structure during integration until the subsequent transfer of the charge packets through the device to be read out. The transfer of this signal charge is, however, not a perfect process. Throughout the lifetime of a space-based mission the detector will be bombarded by high-energy particles and gamma rays. As time progresses, the radiation will damage the detectors, causing the Charge Transfer Efficiency (CTE) to decrease due to the creation of defects or “traps” in the silicon lattice of the detector. The defects create additional energy levels between the valence and conduction band in the silicon of the detector. Electrons or holes (for n-channel or p-channel devices respectively) that pass over the defect sites may be trapped. The trapped electrons or holes will later be emitted from the traps, subject to an emission-time constant related to the energy level of the associated defect. The capture and emission of charge from the signal leads to a characteristic trailing or “smearing” of images that must be corrected to enable the science goals of a mission to be met. Over the past few years, great strides have been taken in the development of the pocket-pumping (or strictly-speaking “trap pumping”) technique. This technique not only allows individual defects (or traps) within the device to be located to the sub-pixel level, but it enables the investigation of the trap parameters such as the emission time constant to new levels of accuracy. Recent publications have shown the power of this technique in characterising a variety of different defects in both n- and p-channel devices and the potential for use in correction techniques, however, we are now exploring not only the trap locations and properties but the life cycle of these traps through time after irradiation. In orbit, most devices will be operating cold to suppress dark current and the devices are therefore cold whilst undergoing damage from the radiation environment. The mobility of defects varies as a function of temperature such that the mix of defects present following a cryogenic irradiation may vary significantly from that found following a room temperature irradiation or after annealing. It is therefore essential to study the trap formation and migration in orbit-like conditions and over longer timescales. In this paper we present a selection of the latest methods and results in the trap pumping of n- and p-channel devices and demonstrate how this technique now allows us to map radiation-induced defects in CCDs through both space and time.

Initial results from a cryogenic proton irradiation of a p-channel CCD

The displacement damage hardness that can be achieved using p-channel charge coupled devices (CCD) was originally demonstrated in 1997 and since then a number of other studies have demonstrated an improved tolerance to radiationinduced CTI when compared to n-channel CCDs. A number of recent studies have also shown that the temperature history of the device after the irradiation impacts the performance of the detector, linked to the mobility of defects at different temperatures. This study describes the initial results from an e2v technologies p-channel CCD204 irradiated at 153 K with a 10 MeV equivalent proton fluences of 1.24×109 and 1.24×1011 protons.cm-2. The number of defects identified using trap pumping, dark current and cosmetic quality immediately after irradiation and over a period of 150 hours after the irradiation with the device held at 153 K and then after different periods of time at room temperature are described. The device also exhibited a flatband voltage shift of around 30 mV per krad, determined by the reduction in full well capacity.

Evolution and Impact of Defects in a p-Channel CCD After Cryogenic Proton-Irradiation

The p-channel charge coupled devices (CCDs) have been shown to display improved tolerance to radiation-induced charge transfer inefficiency when compared with n-channel CCDs. However, the defect distribution formed during irradiation is expected to be temperature dependent due to the differences in lattice energy caused by a temperature change. This has been tested through defect analysis of two p-channel e2v CCD204 devices, one irradiated at room temperature and one at a cryogenic temperature (153 K). Analysis is performed using the method of single trap pumping. The dominant charge trapping defects at these conditions have been identified as the donor level of the silicon divacancy and the carbon interstitial defect. The defect parameters are analyzed both immediately postirradiation and following several subsequent room-temperature anneal phases up until a cumulative anneal time of approximately 10 months. We have also simulated charge transfer in an irradiated CCD pixel using the defect distribution from both the room-temperature and cryogenic case, to study how the changes affect imaging performance. The results demonstrate the importance of cryogenic irradiation and annealing studies, with large variations seen in the defect distribution when compared to a device irradiated at room-temperature, which is the current standard procedure for radiation-tolerance testing.

Postirradiation behavior of p-channel charge-coupled devices irradiated at 153 K

Abstract. The displacement damage hardness that can be achieved using p-channel charge-coupled devices (CCD) was originally demonstrated in 1997, and since then a number of other studies have demonstrated an improved tolerance to radiation-induced charge transfer inefficiency when compared to n-channel CCDs. A number of recent studies have also shown that the temperature history of the device after the irradiation impacts the performance of the detector, linked to the mobility of defects at different temperatures. The initial results from an e2v technologies p-channel CCD204 irradiated at 153 K with 10-MeV equivalent proton fluences of 1.24×109 and 1.24×1011  protons cm−2 is described. The dark current, cosmetic quality, and the number of defects identified using trap pumping immediately were monitored after the irradiation for a period of 150 h with the device held at 153 K and then after different periods of time at room temperature. The device also exhibited a flatband voltage shift of around 30  mV/krad, determined by the reduction in full well capacity.

Evolution of proton-induced defects in a cryogenically irradiated p-channel CCD

P-channel CCDs have been shown to display improved tolerance to radiation-induced charge transfer inefficiency (CTI) when compared to n-channel CCDs. This is attributed to the properties of the dominant charge-trapping defect species in p-channel silicon relative to the operating conditions of the CCD. However, precise knowledge of defect parameters is required in order to correct for any induced CTI. The method of single trap-pumping allows us to analyse the defect parameters to a degree of accuracy that cannot be achieved with other common defect analysis techniques such as deep-level transient spectroscopy (DLTS). We have analysed using this method the defect distribution in an e2v p-channel CCD204 irradiated with protons at cryogenic temperature (153K). The dominant charge trapping defects at these conditions have been identified as the donor level of the silicon divacancy and the carbon interstitial defect. The defect parameters are analysed both immediately post irradiation and following several subsequent room-temperature anneal phases. The evolution of the defect distribution over time and through each anneal phase provides insight into defect interactions and mobility post-irradiation. The results demonstrate the importance of cryogenic irradiation and annealing studies, with large variations seen in the defect distribution when compared to a device irradiated at room-temperature, which is the current standard procedure for radiation testing.

Charge transfer efficiency in a p-channel CCD irradiated cryogenically and the impact of room temperature annealing

It is important to understand the impact of the space radiation environment on detector performance, thereby ensuring that the optimal operating conditions are selected for use in flight. The best way to achieve this is by irradiating the device using appropriate mission operating conditions, i.e. holding the device at mission operating temperature with the device powered and clocking. This paper describes the Charge Transfer Efficiency (CTE) measurements made using an e2v technologies p-channel CCD204 irradiated using protons to the 10 MeV equivalent fluence of 1.24×109 protons.cm-2 at 153 K. The device was held at 153 K for a period of 7 days after the irradiation before being allowed up to room temperature where it was held at rest, i.e. unbiased, for twenty six hours to anneal before being cooled back to 153 K for further testing, this was followed by a further one week and three weeks of room temperature annealing each separated by further testing. A comparison to results from a previous room temperature irradiation of an n-channel CCD204 is made using assumptions of a factor of two worse CTE when irradiated under cryogenic conditions which indicate that p-channel CCDs offer improved tolerance to radiation damage when irradiated under cryogenic conditions.