ASEN-5335 Aerospace Environments -- Radiation Effects on Space Systems 1 MICROELECTRONIC SYSTEMS Emitter-coupled logic Transistor-transistor logic (TTL)/Shotky.

ASEN-5335 Aerospace Environments -- Radiation Effects on Space Systems 1

MICROELECTRONIC SYSTEMS

Emitter-coupled logic

Transistor-transistor logic (TTL)/Shotky TTL

Complementary metal-oxide semiconductor

N-channel transistor

As charge carrier diffusion lengths in electronic devices decrease, they too will cease to function.

Electronic devices rely on the diffusion of charge carriers through semiconducting material to operate properly. We will discuss semiconductor diffusion in more detail in the context of solar cells.

Diffusion lengths for semiconductors (~ 1 m) may be an order of magnitude smaller than those of solar cells, making them especially susceptible to radiation damage.


In addition to electronics, radiation can also degrade the mechanical/electrical/thermal properties of various aerospace polymer films:


DOSE RATE EFFECTS

€

Dρ

IP=

Rad × gcm-3

J=

.01 J Kg-1 × gcm-3

J

D = dose in Rads = density of the materialIP = ionization potential

The number of free electrons produced per unit volume is given by:

In addition to atomic displacements, which decrease the diffusion lengths in electronic devices, radiation will also produce a certain amount of ionizations.

The number of ionizations produced depends on the ionization potential of the material and its density.


Knowing g, the ionization current produced in an np junction by the passage of radiation may be estimated:

€

I = eAWgdD

dt

I = current (amp)e = charge on electron (C)A = area (m2)W = junction width (m)

dD/dt = DOSE RATE

If the DOSE RATE is high enough, the current produced by the radiation may exceed the nominal currents in the device.

Consequently, the nominal signal may be swamped and the device may cease to function properly.

A convenient term is the carrier generation constant

€

g =ρ

IP (electron- hole pairs cm-3 Rad-1)

On average, 1 electron-hole pair is generated per 3.6 eV of energy deposited in a typical semiconductor material like Si.


Many electronic devices have dimensions so small that the currents from the passage of a single energetic particle may be sufficient to alter the operating characteristics of the device in question.

Deposition of energy in an electronic device by a galactic cosmic ray

Production of numerous secondary particles from the nuclear interaction of a primary particle in a sensitive volume of a solid state device.


• Disruptions resulting from single energetic particles are called Single Event Effects (SEE).

• An effect is classified as "soft" if the damage is transitory and recoverable, and as "hard" if the damage is permanent, such as an irreversible bit flip.

• Two specific examples of SEE are latchup (SEL) and upset (SEU).

• NASA defines SEUs to be “radiation-induced errors in microelectronic circuits caused when charged particles (usually from the radiation belts or from cosmic rays) lose energy by ionizing the medium through which they pass, leaving behind a wake of electron-hole pairs.” (SEUs are soft and non-destructive.)

• SEL is when the device is transformed to an anomalous state that no longer responds to input signals. For instance control systems (i.e., propulsion, attitude control, etc.) can be switched into an undesirable mode from which there may be no reset option.

Single Event Effects (SEE)


For typical integrated circuits, the latchup threshold is on the order of 108 Rad/sec.

Upset occurs when a device functions in a manner that is not consistent with its design characteristics. For example, the resulting localized electric fields and currents associated with radiation-induced currents may cause a memory register to change its state.

Upsets are highly device dependent; thresholds can range between 107 and 1012 Rads/sec.

Other common effects associated with upsets include:

• damage to stored data

• damage to software

• central processing unit (CPU) to halt

• the CPU to write over critical data tables


Several approaches for dealing with environmental effects on electronics systems in space, and that are currently in use:

• parts selection -- some older technologies have larger sensitive areas (> 10 m) and smaller upset rate. Tradeoffs are speed, power consumption.

• triple memory redundancy -- tradeoffs are additional weight/power and protection for 'common' electronics.

• error detection and correction -- uses a single memory, but additional stored bits to support the EDAC. Also, the processor and software must run on a regular basis, stealing time (perhaps) from productive work.

• regular reset of on-board computers, or regular re-programming from the ground.

• shielding -- place structures between sensitive components and the environment to minimize dose and dose rate.

Accelerator data indicate that the proton-induced (E > 40 MeV) soft-error rate is < 10-6 soft errors/proton/cm2 . An upper limit to the expected error rate (errors/sec) can therefore be found by using the proton flux models for E > 40 MeV times 10-6.

However, quite often GCR's dominate the soft error rate, and shielding cannot stop GCR’s (or SEP).


http://crsp3.nrl.navy.mil/creme96/



Effects on Aircraft

Aircraft Avionics

Electronic components of aircraft avionic systems are susceptible to damage from the highly ionizing interactions of cosmic rays, solar particles and the secondary particles generated in the atmosphere.

As these components become increasingly smaller, and therefore more susceptible, then the risk of damage also increases. This can corrupt systems leading to erroneous commands.

Problems are expected to increase as more low-power, small-feature size electronics are deployed in “more electric” aircraft. The aviation industry has already catalogued such events on equipment: auto-pilots tripping out and flight instrument units latching into built-in tests.


The left-hand curve gives the number of neutrons that pass through each square centimeter of surface every second (the neutron flux). The right-hand plot gives the cumulative number of memory upsets at an altitude of 30,000 feet after a given number of hours in the air. (Taber and Normand (1993), and published in the IEEE Transactions on Nuclear Science, vol. 40, No. 2, pp 120.)

Left: NASA Altair UAV (100,000 ft. cruise altitude).

High-altitude aircraft are particularly susceptible to SEUs due to neutrons from cosmic ray showers


Earth’s magnetic field acts as a shield and the atmosphere provides an additional barrier to cosmic rays.

As a consequence, the dose rate is dependent on altitude and location.

The background flux is modulated by the solar cycle and by coronal mass ejections.

Intense solar flares can add to the dose rate for short intervals.

Trans-Polar Flights are Increasingly Attractive for Economic Reasons but Imply Increased Exposure to Radiation Harmful to Avionics and to Humans


• What space weather information does the aviation industry need to maintain safe flight operations?

• How do pilots, dispatchers, and air traffic control (ATC) decide when to use a forecast/alert to modify the operations of a flight?

• How much risk are air crew, passengers, and the federal government willing to assume?

• How do we go about educating air crew and passengers about space weather risks during flight?

• How will space weather information be integrated into meteorological information for use by airlines, business jets, Federal Aviation Administration (FAA), ATC, and the commercial space transportation industry?

• Do we need federal laws/regulations in place to ensure the safety of passengers and crew as it pertains to space weather?

• What are the costs/benefits of providing a global space weather service to the aviation industry?


Power Delivered by the Solar Panels on the Solar Heliospheric Observatory at L1

The graph shows a decline by 14% in the first seven years of the satellite's operation. These effects are caused by solar cell degradation due to galactic cosmic rays (GCRs) and energetic solar protons, the topic of this lecture. Solar proton events (SPEs) each reduced performance almost by an additional 2% in each instance, costing the satellite nearly two years (total) of ordinary lifespan. It is expected that solar panels will degrade by about 20-25% during the 10 to 15-year lifetimes of modern GEO satellites. Solar panels are oversized at launch by 25% to allow for this loss of power at the end of the mission lifetime. The problem is that solar flares and proton storms can upset these calculations and cause a satellite to end its service several years earlier, at a large cost to profit margins.

SPE July 14, 2000

SPE Nov 4 & 23, 2001

SOLAR CELL DEGRADATION

Mainly GCR effects here


SOLAR CELL DEGRADATION

Silicon

λ=.35m

.2m

λ=.46m λ=.94m

2.0m200m

Depth ofPenetration

99%atAbsorption

Note: A single photonneedn’t be thisenergetic; absorptionof a continuous spectrum of radiation,absorbed in steps,can release an electron.

As light enters a semiconductor, interactions cause electrons to be released. The depth of penetration of light depends on wavelength:

A SOLAR CELL is a semiconductor device capable of directly converting light energy to electricity. In Si, the ionization potential is 3.6 eV, or a wavelength of .345 m.


The integrated effect of absorbing the whole solar spectrum is to free the numbers of electrons shown in the following graph:

CarrierDensityelectrons/cm3/s

1017

1023

1021

1019

100 200Distance from Front Surface, m

These electrons have been boosted up into the conduction band, leaving a hole (+) behind in the valence band of the bound Si atoms.

These electrons wander aroundthe lattice structure and are“carriers”

In semiconductors and insulators, there is a bandgap above the valence band, followed by a conduction band above that. In metals, the conduction band has no energy gap separating it from the valence band. If electrons in the valence band of a semiconductor are given enough energy, they can jump the gap and enter the conduction band, and thereby conduct electricity.


Left alone, these electron-hole pairs will recombine and liberate thermal energy roughly equivalent to the energy required to free the electrons (analogous to evaporation and condensation of water).

p-type Si200 m

.2 m - n type Si np junction

The SOLAR CELL introduces an additional feature to separate and collect electron - hole pairs before they recombine -- this is an electric field produced in the cell by a np junction, produced by taking an n-type Silicon wafer and adhering it to a p-type Silicon base:


“Empty state” in valence band i.e., nearby electrons jump into void

• In the latter case, for instance, P, As, Sb have 5 valence electrons, but only 4 are needed for the impurity atom to fit into the Si lattice structure; the extra electron is free to roam around at room temperature, and hence becomes the n-type carrier.

• In this context the impurity atom is called a donor, for obvious reasons.

• In the p-type doping, the impurity atom has only 3 valence electrons, and accepts one (hence the name "acceptor") from a nearby Si bond, thus completing its own bonding scheme, and in the process creating a "hole" that wanders around the lattice.

Note:

Barium, Gallium, Indium, Aluminum

• p-type Si is produced by adding doping atoms (Ba, Ga, In, Al) so that the hole concentration (essentially the absence of electrons) is increased.

Phosphorous, Arsenic, Antimony

• n-type Si is produced by adding doping atoms such as P, As, Sb to the Silicon crystal.


However, diffusion is allowed to take place only within certain limits; this is because the diffusing electrons and holes recombine in the vicinity of the junction, producing a net separation of charge, so that an electrostatic field becomes set up to oppose further diffusion (the net flow is now zero).

N-type+ + + + + ++ + + + + ++ + + + + +

P-type- - - - - -- - - - - -- - - - - -

Before forming NP Junction

- - - - - -- - - - - -

+ + + + + ++ + + + + +

- >>+ + >>- (~104)

At the np junction, the electrons have the tendency to diffuse over into the p-type material, since electron concentrations there are low (diffusion occurs in response to a density gradient). Similarly the holes tend to diffuse over to the n-type material for the same reason.

Note: A hole can be treated a just another type of subatomic particle


The equilibrium situation is that a region of carrier depletion ("depletion region" or "space charge region") is formed near the metallurgical np junction, and an electric field exists there in a direction to "put the carriers back where they were" (see figure, following page).

N-type++ + + + +++ + + + +++ + + + +

P-type- - - - --- - - - --- - - - --

After forming NP Junction

- - - - - - - -

+ + + + + + + +

space charge region

In the operation of a SOLAR CELL, the absorption of radiation liberates electrons -- these are either recaptured (thereby releasing heat), or, if they reach the np junction, will be accelerated into the n-type material.


0.1 1.0 10 100 200Distance from Front Surface, m

100104108

10121016

CARRIERDENSITY

CM-3

n p depletion region

-E field


Ee-

~ 1 Voltfew mA

LIGHT

SCHEMATIC OF SINGLE SOLARCELL OPERATION

Once in the n-type material, there are few atoms capable of capturing the electron -- it is free to move to the surface of the cell. If electrical leads are placed on the surface, this flow of electrons can be drawn off as current flow, which in turn can be used to power a spacecraft.

- +

light


A single cell generates a potential difference of about 1 Volt and a current of a few mA. By connecting the cells in series, a larger potential difference can be generated; by increasing the number of "strings" of cells, additional current can be generated.

The operation of a SOLAR CELL depends on the ability of an electron to diffuse to the depletion region before it is captured by a hole; processes which shorten its mean free path or its "lifetime", will degrade cell performance.

A shorter mean free path lengthens diffusion time to the depletion region.

HOW DOES PARTICLE RADIATION DEGRADE SOLAR CELL PERFORMANCE?

Typical efficiencies for silicon solar cells are ~20% (that is, ~80% of incident solar energy is converted to heat).


As energetic particles interact with a solar cell, they produce ionizations and atomic displacements.

If an Si atom is displaced from the lattice structure, the structure of the lattice is altered (forming a “stable defect”), effectively decreasing the distance an electron can diffuse before being deflected by the lattice defect.

The MAGION microsatellites were launched by the Czech Republic. This figure shows MAGION-4 and MAGION-5 solar array degradation during the 25 months after launch. Most of the differences between these two curves can be explained by the effect of the radiation belts, which is more than 10 times stronger along the orbit of MAGION-5 than MAGION-4. The solar array degradation curve measured on the low-orbiting MAGION-2 (apogee 2400 km) is shown for comparison. (Triska et al., 2005).

Consequently both the current and voltage drop of a solar cell (and hence power) are decreased:


• To determine the effect of radiation on a particular solar cell, it is necessary to know both the differential energy spectrum of the radiation, and the response characteristics of the cell.

• The differential fluence is usually converted into the equivalent fluences of either 1 MeV electrons, or 10 MeV protons; that is, the damage produced by a distributed spectrum of particles is equated to a number of mono-energetic particles that would be required to produce the same damage.

• One 10 MeV proton produces the same damage as about 3000 1 MeV electrons.

• Besides Si, solar cells can also be formed by GaAs and InP.

• While these technologies offer superior radiation resistance and increased efficiency, additonal manufacturing costs require the designer to perform a system level trade to determine which technology is most appropriate for the orbital environment in question.


Solar Proton Effects on Solar ArraysAt geosynchronous orbit the trapped radiation environment consists of electrons and protons.

– There are very few trapped protons > 10 MeV in the geosynchronous region.

– Normally solar array cells are covered with thin glass covers to shield the cells from all protons < 10 MeV.

Solar particle events can generate high fluences of > 10 MeV protons in relatively short periods (days).

– The fluence of > 50 MeV protons can be substantial.

– These protons increase the solar array damage significantly, reducing the mission life.

The GOES-7 solar array current was reduced by nearly 10% as a result of two large solar events in 1989 (see following figure).


GOES 7 Solar Array Current Degradation

ASEN-5335 Aerospace Environments -- Radiation Effects on Space Systems 1 MICROELECTRONIC SYSTEMS Emitter-coupled logic Transistor-transistor logic (TTL)/Shotky.

Documents

radiation effects

radiation damage

passage of radiation

radiation belts

space systems

aerospace environments

radiationinduced errors

electronic devices