Space-RHA
Tools / Dose and damage / NIEL conversion tool

Proton NIEL and displacement damage dose

Proton NIEL and displacement-damage calculations using the Jun et al. 2003 dataset. Look up NIEL, convert fluence between energies, calculate DDD, build compound materials with Bragg's rule, or convert to 1 MeV neutron equivalent fluence.

v1.0 · Apr 2026Jun et al. 2003 NIEL tables10 elements, any compound by Bragg's rule0.08 keV to 1000 MeVruns in your browserHelp & Reference

Material

Energy

Valid range: 0.08 keV to 1000 MeV (material-dependent)
NIEL of the selected material against proton energy on log-log axes, through the tabulated points of Jun et al. 2003; the marker is the energy you looked up.
NIEL at selected energy
--
MeV-cm²/g
Equivalent in keV-cm²/g
--
Damage energy cross section (MeV-mb)
--

Material

Fluence Conversion

The same displacement damage at two energies: the fluence scales by the ratio of the NIEL at the source energy to the NIEL at the target energy.
NIEL at source energy
--
MeV-cm²/g
NIEL at target energy
--
MeV-cm²/g
NIEL ratio (source/target)
--
Equivalent fluence at target energy
--
p/cm²
Φtarget = Φsource × NIEL(Esource) / NIEL(Etarget)

Material

Displacement Damage Dose (DDD)

The dose is the fluence times the NIEL at the proton energy, read from the curve at the marker.
NIEL at energy
--
MeV-cm²/g
Displacement Damage Dose
--
MeV/g
DDD = Φ × NIEL(E)

Material

1 MeV neutron equivalent fluence

Reference: NIEL(1 MeV neutron, Si) = 1.95 × 10-3 MeV-cm²/g
The 95 MeV-mb reference of ASTM E722 converts to 2.04 × 10-3 MeV-cm²/g, so this result is about 4.5 percent higher than that value gives.
The hardness factor is how far the marker sits above or below the dashed line, the NIEL of a 1 MeV neutron in silicon.
NIEL at proton energy (in material)
--
MeV-cm²/g
Displacement Damage Dose
--
MeV/g
1 MeV Neutron Equivalent Fluence (Si reference)
--
neq/cm²
Hardness factor (NIELp / NIEL1MeV n,Si)
--
Φneq = Φp × NIELp(E) / NIEL1MeV n,Si

Materials to plot

NIEL vs Proton Energy (Log-Log)

Data from Jun et al. IEEE TNS 50(6), 2003. Log-log interpolation between data points.

Material

All values in MeV-cm²/g, every row of Table I of the paper for each element (silicon 36 rows, 0.2 keV to 1000 MeV); re-transcribed against the published table on 29 September 2026.

NIEL Data Table (Jun et al. Table I)

How it works

Calculating displacement damage dose

01

Pick the material

Choose one of the ten elements in the Jun et al. tables, or build a compound from its formula: Bragg's rule weights each element's NIEL by its mass fraction.

02

Read NIEL at the energy

NIEL is interpolated log-log between the tabulated energies that bracket yours. Outside a material's table the tool says out of range rather than extrapolating.

03

Scale by the fluence

The fluence times the NIEL is the displacement damage dose. A ratio of two NIELs moves a fluence to another energy, and scaling the fluence by its NIEL over the 1 MeV neutron NIEL in silicon gives the neutron equivalent.

Understanding the mechanism

Understanding displacement damage: what one particle does to the lattice, and why only some devices care

The calculators above turn a fluence into a displacement damage dose. The model below shows what that dose is made of. A particle knocks a silicon atom off its site; the recoil shares its energy between ionization, which displaces nothing, and further collisions, which leave vacancies and interstitials, isolated for electrons and in dense clusters for neutrons and the nuclear recoils from protons. Most of those defects recombine within seconds at room temperature and the clusters restructure for minutes; the survivors are deep levels. Devices that depend on minority-carrier lifetime, on generation in a depletion region or on a lightly doped bulk feel them; a heavily doped surface-channel CMOS transistor generally does not. Fire a particle, anneal, then irradiate six devices to the same dose.

The kinematics, the Lindhard partition in Robinson's form, the NRT displacement count, the silicon proton and electron NIEL tables of Jun and co-workers (2003, 2009), the Messenger and Spratt gain form, the Srour and Lo dark-current factor and the ASTM neutron reference are used as published; every other constant is illustrative, chosen so that the direction and rough size of each effect follows the papers listed below. The lattice is drawn to scale as a block of three cells a side; the cascade drawing is schematic, since a 100 keV recoil travels about 200 lattice constants. The model cannot predict a part, and its device curves are orderings, not design margins.

Where each effect in the model comes from

The collision and the cascade. Recoil energies follow two-body kinematics; the partition of a recoil between ionization and atomic motion is Lindhard's energy partition function, applied in Burke's first complete silicon proton NIEL [3] and, in Robinson's analytical form, in the JPL and NRL tabulations [17], [18] and the NEMO code [19] (the Geant4 module of Weller and co-workers [19] tracks the cascade instead); NIEL is a damage-energy rate, and the NRT displacement count the model shows is the convention for turning it into displaced atoms [17]. The silicon NIEL the model quotes is Jun and co-workers 2003 [18], read from the paper's table with its displacement threshold of 21 eV, and the electron NIEL is Jun and co-workers 2009 [18]; the threshold choice moves proton NIEL by about 15 percent and electron NIEL below 1 MeV by 40 percent or more [19], and Inguimbert 2024 [24] finds that electron damage data are fitted best by an effective threshold near 77 eV. That damage per unit NIEL depends on the recoil energy is the electron result of Dale and co-workers 1988 [7] and Summers and co-workers 1993 [8], framed by the point-defect against cluster picture of Srour and Palko 2006 [21], with the injection-level dependence that goes with it: per unit damage energy, generation in the cluster regime is 45 to 76 times that of the isolated-defect regime, while among protons and neutrons of every energy the mean damage is the same per unit NIEL [10]; the effective NIEL that weights recoil energy by the defects that survive the cascade is Inguimbert and co-workers 2010 [22]; the individual cascades and their relaxation, which is what any DDD averages over, are Auden and co-workers 2012, Raine and co-workers and Jay and co-workers 2017 [23]. The boundary the model draws at about 2 keV of damage energy is Wood's grouping as quoted in [5], [8] and [23] (free defects below 1 to 2 keV, one cascade to 12 keV, subcascades above); Srour and Palko fit a lower 300 eV for the first clustering [21]. The surviving fractions the model keeps (15 percent of point defects, 30 percent of cluster displacements) are illustrative: more than 90 percent of Frenkel pairs recombine [8], [10], about a fifth of a large pocket remains after a minute in molecular dynamics [23], and the annealing behavior of the stable complexes is reviewed in [1] and [2].

NIEL and DDD. The correlation of device damage with calculated NIEL for protons across energies is Burke 1986 [3] with the companion bipolar measurements of Summers and co-workers [4] (5 MeV protons 8.5 times and 60 MeV protons 1.8 times as damaging as fission neutrons), extended across protons, deuterons, helium ions and neutrons by Summers and co-workers 1987 [5], who state the 1 MeV neutron reference as 2.04e-3 MeV cm2/g (95 MeV mb; this tool's 1.95e-3 is explained on the help page), and to gallium arsenide by Burke and co-workers [6], who also saw the first shortfall at 187 MeV where energetic recoils leave a thin implanted layer. The displacement damage dose itself, fluence times NIEL, is the metric of Dale and co-workers 1989 (outside TNS) and the canonical reference is Summers and co-workers 1993 [8]. Electrons are device dependent: in n-type silicon and n-type gallium arsenide the electron damage coefficient is linear in NIEL and on the proton line [8], in bipolar transistors of both base types it is linear with the lowest-energy electrons up to twice as damaging as the line predicts [7], and in p-type silicon solar cells it is quadratic, about ten times below the n-type value at 1 MeV [8], so a solar-cell electron dose needs an effective-DDD correction, an exponent between 1 and 3 fitted to the device, formalized for on-orbit use by Inguimbert and Messenger 2012 [22]. The model applies an exponent of 1.7 (the value Inguimbert 2024 [24] calls common; [22] gives 2.4 for silicon maximum power) to the solar cell and to the imager, and none to the bipolar transistor, the LED and the MOSFETs; its anchoring of the electron curve to the proton line at the electron NIEL near 100 MeV, where the electron recoil spectrum resembles that of low-energy protons [7], is a model choice and labeled so. Where NIEL scaling holds is Messenger and co-workers 2004 [20]: the incident particle must outrange the active volume (silicon solar cells depart below about 8 MeV, gallium arsenide cells below 0.3 to 0.4 MeV), the energetic recoils must stay inside it, and enough interactions must occur for the average to mean anything; protons that stop in the active volume are handled by the adjusted NIEL of Messenger and co-workers 2002 [16]. Heavy-ion NIEL is published (Messenger and co-workers 1999 and 2003 [17]: iron on silicon about 2 to 3 MeV cm2/g at 1 GeV) and fission-fragment damage follows it in a thick diode [23]; this site carries no heavy-ion table. The uncertainty in the partition function and in the threshold energy, which propagates into every tabulated NIEL including the one in this tool, is Griffin and Cooper 2017 (about 15 percent) and Inguimbert 2024 [24].

Devices. Which electrical property a displacement defect degrades, and therefore which devices are sensitive, follows the reviews [1], [2]: minority-carrier lifetime for bipolar transistors, solar cells and light emitters; generation current in depleted volumes for imagers and detectors; carrier removal in lightly doped bulk for power devices; and, only at much higher dose, mobility in majority-carrier channels. The bipolar gain form is Messenger and Spratt 1958 (outside TNS); the solar cell characteristic curve is the NRL DDD method of Messenger and co-workers 2001 (outside TNS), drawn here with illustrative constants for a silicon cell; the dark-current factor per unit DDD is Srour and Lo 2000 [12], the mean over many device types with a spread of about 30 percent, valid for bulk generation in the proton and neutron regime one week after irradiation, used with a pixel volume the visitor chooses. The lifetime damage rate, the carrier-removal rate and every device constant in the panel are illustrative orderings; the reviews give the measured ranges.

Pixel statistics. That the mean dark current follows the DDD while the distribution across pixels and its hot-pixel tail are set by the variance of the energy deposited per microvolume, with single proton nuclear interactions making the extremes, is Srour, Hartmann and Kitazaki 1986 [9] and the four Marshall and Dale papers of 1989 and 1990 [10]: a Poisson number of nuclear events per pixel (cross sections of 0.5 to 0.9 barn per silicon atom for 12 to 63 MeV protons, about 3 barn for fission neutrons), each depositing several hundred times the damage energy of a Coulomb recoil, on top of a Coulomb core whose relative width falls with the number of events. The model's Poisson-plus-exponential picture uses those cross sections and the nuclear share of the dose from [10]. A second kind of hot pixel, which the model does not draw, comes from field-enhanced emission at defects in high-field regions, even from low-energy recoils: Srour and Hartmann 1989 [11] measured its lowered activation energies, and Marshall and Dale [10] found the extremes three times larger in the higher-field design of the same imager. An early end-to-end DDD prediction for a shielded flight imager, giving 3 to 9 million MeV/g per year in polar low Earth orbit, is Dale and co-workers 1993 [13]; the DDD against shielding depth curves, for primary protons, are Messenger and co-workers 1997 [14]; the secondary particles from proton interactions along the path in shielding are quantified by Jun and McAlpine 2001 [15], who find neutrons the only ones that matter (pions, deuterons and alphas add at most 5 percent), taking over from the primaries beyond about 10 to 16 g/cm2 of aluminum; and the extension of DDD to protons that stop in the active volume is Messenger and co-workers 2002 [16].

The reviews [1] and [2] are the entry points for everything above. Origins outside TNS named in the text: G. C. Messenger and J. P. Spratt, Proc. IRE 46, 1958; J. Lindhard et al., Mat. Fys. Medd. 33, 1963; M. J. Norgett, M. T. Robinson and I. M. Torrens, Nucl. Eng. Des. 33, 1975; C. J. Dale, P. W. Marshall, G. P. Summers, E. A. Wolicki and E. A. Burke, Appl. Phys. Lett. 54, 451, 1989; S. R. Messenger, G. P. Summers, E. A. Burke, R. J. Walters and M. A. Xapsos, Prog. Photovolt. 9, 103, 2001.

  1. J. R. Srour, C. J. Marshall, P. W. Marshall, "Review of Displacement Damage Effects in Silicon Devices," IEEE Trans. Nucl. Sci. 50(3), p. 653, 2003.
  2. J. R. Srour, J. W. Palko, "Displacement Damage Effects in Irradiated Semiconductor Devices," IEEE Trans. Nucl. Sci. 60(3), p. 1740, 2013.
  3. E. A. Burke, "Energy Dependence of Proton-Induced Displacement Damage in Silicon," IEEE Trans. Nucl. Sci. 33(6), p. 1276, 1986.
  4. G. P. Summers, E. A. Wolicki, M. A. Xapsos, P. Marshall, C. J. Dale, M. A. Gehlhausen, R. D. Blice, "Energy Dependence of Proton Displacement Damage Factors for Bipolar Transistors," IEEE Trans. Nucl. Sci. 33(6), p. 1282, 1986.
  5. G. P. Summers, E. A. Burke, C. J. Dale, E. A. Wolicki, P. W. Marshall, M. A. Gehlhausen, "Correlation of Particle-Induced Displacement Damage in Silicon," IEEE Trans. Nucl. Sci. 34(6), p. 1134, 1987.
  6. E. A. Burke, C. J. Dale, A. B. Campbell, G. P. Summers, et al., "Energy Dependence of Proton-Induced Displacement Damage in Gallium Arsenide," IEEE Trans. Nucl. Sci. 34(6), p. 1220, 1987.
  7. C. J. Dale et al., "High Energy Electron Induced Displacement Damage in Silicon," IEEE Trans. Nucl. Sci. 35(6), p. 1208, 1988.
  8. G. P. Summers, E. A. Burke, P. Shapiro, S. R. Messenger, R. J. Walters, "Damage Correlations in Semiconductors Exposed to Gamma, Electron and Proton Radiations," IEEE Trans. Nucl. Sci. 40(6), p. 1372, 1993.
  9. J. R. Srour, R. A. Hartmann, K. S. Kitazaki, "Permanent Damage Produced by Single Proton Interactions in Silicon Devices," IEEE Trans. Nucl. Sci. 33(6), p. 1597, 1986.
  10. P. W. Marshall, C. J. Dale, E. A. Burke, G. P. Summers, G. E. Bender, "Displacement Damage Extremes in Silicon Depletion Regions," IEEE Trans. Nucl. Sci. 36(6), p. 1831, 1989; C. J. Dale, P. W. Marshall, E. A. Burke, G. P. Summers, G. E. Bender, "The Generation Lifetime Damage Factor and Its Variance in Silicon," IEEE Trans. Nucl. Sci. 36(6), p. 1872, 1989; P. W. Marshall, C. J. Dale, E. A. Burke, "Proton-Induced Displacement Damage Distributions and Extremes in Silicon Microvolumes," IEEE Trans. Nucl. Sci. 37(6), p. 1776, 1990; C. J. Dale, P. W. Marshall, E. A. Burke, "Particle-Induced Spatial Dark Current Fluctuations in Focal Plane Arrays," IEEE Trans. Nucl. Sci. 37(6), p. 1784, 1990.
  11. J. R. Srour, R. A. Hartmann, "Enhanced Displacement Damage Effectiveness in Irradiated Silicon Devices," IEEE Trans. Nucl. Sci. 36(6), p. 1825, 1989.
  12. J. R. Srour, D. H. Lo, "Universal Damage Factor for Radiation-Induced Dark Current in Silicon Devices," IEEE Trans. Nucl. Sci. 47(6), p. 2451, 2000.
  13. C. J. Dale, P. W. Marshall, B. Cummings, L. Shamey, A. Holland, "Displacement Damage Effects in Mixed Particle Environments for Shielded Spacecraft CCDs," IEEE Trans. Nucl. Sci. 40(6), p. 1628, 1993.
  14. S. R. Messenger, M. A. Xapsos, E. A. Burke, R. J. Walters, G. P. Summers, "Proton Displacement Damage and Ionizing Dose for Shielded Devices in Space," IEEE Trans. Nucl. Sci. 44(6), p. 2169, 1997.
  15. I. Jun, W. McAlpine, "Displacement Damage in Silicon Due to Secondary Neutrons, Pions, Deuterons, and Alphas From Proton Interactions With Materials," IEEE Trans. Nucl. Sci. 48(6), p. 2034, 2001.
  16. S. R. Messenger, E. A. Burke, G. P. Summers, R. J. Walters, "Application of Displacement Damage Dose Analysis to Low-Energy Protons on Silicon Devices," IEEE Trans. Nucl. Sci. 49(6), p. 2690, 2002.
  17. S. R. Messenger, E. A. Burke, G. P. Summers, M. A. Xapsos, R. J. Walters, E. M. Jackson, B. D. Weaver, "Nonionizing Energy Loss (NIEL) for Heavy Ions," IEEE Trans. Nucl. Sci. 46(6), p. 1595, 1999; S. R. Messenger, E. A. Burke, M. A. Xapsos, G. P. Summers, R. J. Walters, I. Jun, T. Jordan, "NIEL for Heavy Ions: An Analytical Approach," IEEE Trans. Nucl. Sci. 50(6), p. 1919, 2003.
  18. I. Jun, M. A. Xapsos, S. R. Messenger, E. A. Burke, R. J. Walters, G. P. Summers, T. Jordan, "Proton Nonionizing Energy Loss (NIEL) for Device Applications," IEEE Trans. Nucl. Sci. 50(6), p. 1924, 2003; I. Jun, M. A. Xapsos, E. A. Burke, "Alpha Particle Nonionizing Energy Loss (NIEL)," IEEE Trans. Nucl. Sci. 51(6), p. 3207, 2004; M. A. Xapsos, E. A. Burke, F. F. Badavi, L. W. Townsend, J. W. Wilson, I. Jun, "NIEL Calculations for High-Energy Heavy Ions," IEEE Trans. Nucl. Sci. 51(6), p. 3250, 2004; I. Jun, W. Kim, R. Evans, "Electron Nonionizing Energy Loss for Device Applications," IEEE Trans. Nucl. Sci. 56(6), p. 3229, 2009.
  19. R. A. Weller, M. H. Mendenhall, D. M. Fleetwood, "A Screened Coulomb Scattering Module for Displacement Damage Computations in Geant4," IEEE Trans. Nucl. Sci. 51(6), p. 3669, 2004; C. Inguimbert, R. Gigante, "NEMO: A Code to Compute NIEL of Protons, Neutrons, Electrons, and Heavy Ions," IEEE Trans. Nucl. Sci. 53(4), p. 1967, 2006.
  20. S. R. Messenger, E. A. Burke, G. P. Summers, R. J. Walters, "Limits to the Application of NIEL for Damage Correlation," IEEE Trans. Nucl. Sci. 51(6), p. 3201, 2004.
  21. J. R. Srour, J. W. Palko, "A Framework for Understanding Displacement Damage Mechanisms in Irradiated Silicon Devices," IEEE Trans. Nucl. Sci. 53(6), p. 3610, 2006.
  22. C. Inguimbert, P. Arnolda, T. Nuns, G. Rolland, "Effective NIEL in Silicon: Calculation Using Molecular Dynamics Simulation Results," IEEE Trans. Nucl. Sci. 57(4), p. 1915, 2010; C. Inguimbert, S. Messenger, "Equivalent Displacement Damage Dose for On-Orbit Space Applications," IEEE Trans. Nucl. Sci. 59(6), p. 3117, 2012.
  23. E. C. Auden, R. A. Weller, M. H. Mendenhall, R. A. Reed, R. D. Schrimpf, N. C. Hooten, M. P. King, "Single Particle Displacement Damage in Silicon," IEEE Trans. Nucl. Sci. 59(6), p. 3054, 2012; M. Raine, A. Jay, N. Richard, V. Goiffon, S. Girard, M. Gaillardin, P. Paillet, "Simulation of Single Particle Displacement Damage in Silicon, Part I," IEEE Trans. Nucl. Sci. 64(1), p. 133, 2017; A. Jay, M. Raine, N. Richard, N. Mousseau, V. Goiffon, A. Hemeryck, P. Magnan, "Simulation of Single Particle Displacement Damage in Silicon, Part II," IEEE Trans. Nucl. Sci. 64(1), p. 141, 2017.
  24. P. J. Griffin, P. J. Cooper, "Influence of the Damage Partition Function on the Uncertainty of the Silicon Displacement Damage Metric," IEEE Trans. Nucl. Sci. 64(1), p. 574, 2017; C. Inguimbert, "Damage Energy Threshold Anisotropy in Nonionizing Energy Loss Calculation," IEEE Trans. Nucl. Sci. 71(8), p. 1470, 2024.

Which devices are sensitive: displacement damage by device class, from current data

The chart many programs still cite for this is a bar chart of neutron hardness by device technology compiled by M. Rose of Physitron in 1990 and reproduced in the JPL MMIC reliability guideline as Fig. 10-13 [2], [3]. It gives one bar per technology, and no data, test criteria or method behind the bars could be found. The chart below rebuilds it from published test data: for each device class, the lowest fluence at which a measurable change was reported, and the range over which parts failed or changed by roughly 20 to 50 percent or more, each value traced to a paper in the table below. The 1990 bars are drawn in outline for comparison. What orders the classes is generally the minority carrier lifetime: devices that depend on it (bipolar transistors, most of all those with low fT and wide bases, LEDs, optocouplers, image sensors) degrade first, and majority carrier devices (MOSFETs, JFETs, CMOS logic) last, which the Messenger and Spratt relation expresses as a gain loss inversely proportional to fT [1]. Failure is application dependent, and within a class the spread between part types is often two orders of magnitude, so a class bar is a prior, not a rating.

Displacement damage sensitivity by device classFor each device class, a bar for the fluence range over which parts failed or changed by roughly 20 to 50 percent or more, a diamond at the lowest fluence at which a change was reported, and an outlined bar for the 1990 chart. Silicon classes in 1 MeV neutron equivalent fluence in silicon; compound semiconductor classes in the particle reported.1e81e91e101e111e121e131e141e151e16fluence, per cm2Silicon devices: 1 MeV neutron equivalent fluence in siliconSignal and rectifier diodes1990 chart (Rose, Physitron), row SIGNAL DIODES: from 5.4e13, drawn to the right edge of the original chart at 1e15Failure or large change: forward voltage doubles at 1e13 in power rectifiers; switching diode forward voltage changes at 1e14 to 5e14 (fluence above 10 keV). [6]*Lowest fluence with a reported change: 1e11, reverse current rises in rectifiers rated above 10 A (fluence above 10 keV, as compiled). [6]*Zener and reference diodes1990 chart (Rose, Physitron), row REFERENCE DIODES: from 4.3e12, drawn to the right edge of the original chart at 1e15Failure or large change: reverse current up 10 times at 2e14; Zener voltage drifts beyond about 1e14 to 1e15. [6]*Lowest fluence with a reported change: 1e14, Zener voltage still within 1 to 2 percent (fluence above 10 keV). [6]*Low-fT bipolar transistors1990 chart (Rose, Physitron), row LO FT TRANSISTORS: from 9.8e10, drawn to the right edge of the original chart at 1e15Failure or large change: 50 percent gain loss at 4e12 to 5e12 in vertical NPN and PNP transistors and 5e12 to 7e12 in 2N2222A and 2N2907A (derived), about 1.6e12 extrapolated for a 4N49 phototransistor; a lateral PNP loses 20 to 33 percent at 5e11, more at low emitter-base bias. [7], [8], [9]Lowest fluence with a reported change: 3.1e10, about 2 percent gain loss in the phototransistor of a 4N49 (fT about 300 MHz); 10 percent at 4.4e11 to 7.4e11 in dielectrically isolated vertical transistors. [7], [8]High-fT bipolar transistors (microwave)1990 chart (Rose, Physitron), row HI FT TRANSISTORS: from 9.8e12, drawn to the right edge of the original chart at 1e15Failure or large change: about 40 percent gain loss in an IC NPN transistor near 8e13 (800 MeV protons). [10]*Lowest fluence with a reported change: 1e13, microwave transistors show no gain degradation below about 1e13 neutrons (as cited in the JPL chapter). [2]Bipolar power transistors1990 chart (Rose, Physitron), row POWER TRANSISTORS (BIPOLAR): from 9.8e10, drawn to the right edge of the original chart at 1e15Failure or large change: 50 percent gain loss in individual units from 8e10 to about 1e13; type medians 1.3e11 (2N3772) to 7e12 (2N3740). [11]Lowest fluence with a reported change: 9e9, 10 percent gain loss at 9e9 to 1.6e10 in 2N3772 (fT about 1.4 MHz; fast burst reactor fluence). [11]SiGe HBTsFailure or large change: 20 to 88 percent gain loss at 9.9e13 depending on the generation; minimal for the 0.12 um 8HP. [12]Lowest fluence with a reported change: 1.4e13, 5 to 28 percent gain loss (63 MeV protons; part of it ionization at the emitter-base spacer). [12]IGBTsFailure or large change: on-state voltage up 2.2 times and saturation current halved at 1e13. [13]Lowest fluence with a reported change: 1e12, saturation current down 15 to 20 percent. [13]JFETs1990 chart (Rose, Physitron), row JFET TRANSISTORS: from 9.8e13, drawn to the right edge of the original chart at 1e15Lowest fluence with a reported change: 6.5e12, gate leakage up about 1000 times in detector-grade JFETs (27 MeV protons, authors' 1 MeV equivalent). [14]No significant change reported up to 1e14: transconductance unchanged to 2.9e13; minimal gain change reported above 1e14. [10]*, [14]Si power MOSFETs, 150 to 500 V1990 chart (Rose, Physitron), row MOSFET TRANSISTORS: from 9.8e13, drawn to the right edge of the original chart at 1e15Failure or large change: on-resistance up 2 to 24 times at 5.4e14 in 150 to 250 V parts. [16]Lowest fluence with a reported change: 8e10, subthreshold leakage changes in an IRF640 (63 MeV neutrons); on-resistance up 10 to 30 percent at 1.1e13 to 3.3e13 in 400 to 500 V parts. [15], [16]Si power MOSFETs, 30 to 100 VFailure or large change: on-resistance up 1.2 to 4.7 times at 2.8e15. [16]Lowest fluence with a reported change: 1e15, on-resistance up about 20 percent. [16]Linear bipolar ICs1990 chart (Rose, Physitron), row LINEAR ICs (Si) (BIPOLAR): from 5.1e11, drawn to the right edge of the original chart at 1e15Failure or large change: LM137 catastrophic failure at 2.4e11 to 4.9e11; RH1056 at 3e11 to 1.2e12; LM124 input current up 132 nA at 1.7e12; narrow-base processes unaffected to 2.2e12. [17], [18]Lowest fluence with a reported change: 1.1e11, LM137 start-up and LM117 output shifts at 5e10 protons (50 MeV). [17]Digital bipolar ICs1990 chart (Rose, Physitron), row DIGITAL ICs (Si) (BIPOLAR): from 5.1e12, drawn to the right edge of the original chart at 1e15Failure or large change: I2L from 3e13, TTL 6e13 to 2e15, ECL 2.6e14 to 6e15, oxide-isolated logic 6e14 to 6e15 (failure ranges). [4], [5]Lowest fluence with a reported change: 6e12, failures of low-power TTL from about 6e12. [4], [5]CMOS ICs1990 chart (Rose, Physitron), row ALL MOS (Fig. 10-14): from 1.1e14, drawn to the right edge of the original chart at 1e15Failure or large change: n-MOS 1.1e14 to 1e15, bulk CMOS 2.4e14 to 2.2e15, CMOS/SOS 1.3e15 to 1e16 (failure ranges). [4], [5]Lowest fluence with a reported change: 8e10, SDRAM retention time falls from 3 s to 0.2 to 0.6 s (63 MeV neutrons). [15]Si photodiodes and phototransistorsFailure or large change: phototransistor gain down 40 to 60 percent at 5e11 and 60 to 80 percent at 1e12, at low photocurrent. [19]Lowest fluence with a reported change: 1.9e10, 4N49 phototransistor photoresponse down 10 percent (200 MeV protons). [7]CCDs (charge transfer)Failure or large change: significant charge transfer losses below 1 krad(Si) of protons, about 9e9; inefficiency 2.4e-5 at 1.2e10. [20]*, [21]*Lowest fluence with a reported change: 5e8, parallel charge transfer inefficiency up 1e-6 (derived from the damage constant of a science CCD, 10 MeV protons). [20]*CMOS image sensors (pinned photodiode)Failure or large change: quantum efficiency and full well degraded at 6.2e14. [22]Lowest fluence with a reported change: 6.2e9, mean dark current and random telegraph signal rise at the lowest fluence tested. [22]No significant change reported up to 1.9e12: quantum efficiency, conversion gain and full well unchanged to 1.9e12. [22]Optocouplers, amphoteric GaAs:Si LEDFailure or large change: 50 percent CTR loss at 1e10 to 1.8e10 protons (50 MeV) and 1.7e10 to 2.8e10 (63 MeV, low LED current). [7], [23], [24]Lowest fluence with a reported change: 4.9e9, 10 percent CTR loss at about 2.5e9 protons (63 MeV). [23]Optocouplers, AlGaAs double heterojunction LEDFailure or large change: 50 percent loss at 1.6e11 to 3.4e11 protons (50 MeV) and 4.7e11 to 6e11 neutrons. [24], [25]Lowest fluence with a reported change: 4.4e10, 6 to 15 percent loss at the lowest level tested (50 MeV protons); 10 percent at 8.5e10 to 1.3e11 neutrons. [24], [25]Optocouplers, GaAsP LEDFailure or large change: 50 percent loss at 3.3e12 to 3.4e12 neutrons, 90 percent at 6.5e12. [25]Lowest fluence with a reported change: 3.1e10, 3.5 percent loss in a 6N140 (155 MeV protons); 15 percent at 1.9e11. [7]Optocouplers, all LED types: site records1990 chart (Rose, Physitron), row OPTICAL ISOLATORS: from 2.7e10, drawn to the right edge of the original chart at 1e15Site records, 50 percent CTR loss: 5.3e11 at the median, 2.9e10 to 5.3e12 from the 90th to the 10th percentile part type (19 retention statements, 14 public). [33]10 percent CTR loss for the most sensitive decile of part types: 3.2e9. [33]Unijunction transistors (1990 chart only)1990 chart (Rose, Physitron), row UNIJUNCTION TRANSISTORS: from 5.8e11, drawn to the right edge of the original chart at 1e15SCRs (1990 chart only)1990 chart (Rose, Physitron), row SILICON-CONTROLLED RECTIFIERS: from 9.3e10, drawn to the right edge of the original chart at 1e15Compound semiconductor devices: fluence of the particle reported, not convertedLEDs, amphoteric GaAs:Si (50 MeV p)Failure or large change: 50 percent light loss, depending on bias, lot and current. [7], [26]Lowest fluence with a reported change: 2e9, about 10 percent light loss. [26]LEDs, AlGaAs double heterojunction (50 MeV p)Failure or large change: 50 percent light loss (50 to 58 MeV) for typical units. [26], [27]Lowest fluence with a reported change: 5e9, 2 to 3 percent light loss. [26]LEDs, GaAsP (50 MeV p)Lowest fluence with a reported change: 1.5e11, change visible; stated to be more than 10 times harder than amphoteric LEDs. [26]Laser diodes (50 MeV p)Failure or large change: threshold current up 20 to 30 percent, within a factor of 2 across six laser types. [28]Lowest fluence with a reported change: 3e11, threshold current up 2 to 7 percent. [28]GaAs ICs and MMICs (neutrons)1990 chart (Rose, Physitron), row GaAs ICs: from 9.3e13, drawn to the right edge of the original chart at 1e15Failure or large change: JFET transconductance down 20 percent from about 5e13 in lightly doped channels. [2]Lowest fluence with a reported change: 4e13, FET pinch-off voltage starts to fall. [2]No significant change reported up to 1e15: FET small-signal gain holds until near 1e15. [2]GaN HEMTs (neutrons; 10 MeV p)Failure or large change: 20 to 25 percent drain current and transconductance loss at 1e14 protons (10 MeV). [30]*Lowest fluence with a reported change: 6e12, drain current changes at 6e12 neutrons (1 MeV equivalent, compiled onset). [29]SiC power JFET (1 MeV n eq)Failure or large change: on-resistance 2.5 times at 6.6e13, 14.7 times at 1.7e14, non-functional above 4e14. [31]Lowest fluence with a reported change: 1.3e13, on-resistance up 30 percent (authors' 1 MeV equivalent). [31]InP HBTs (neutrons, 1 MeV eq)Failure or large change: gain below 20 (from about 200) at 2e14 to 6e14. [32]Lowest fluence with a reported change: 2.3e13, gain already halved at the lowest exposure. [32]1e81e91e101e111e121e131e141e151e16failure or large change (roughly 20 to 50 percent or more)lowest fluence with a reported changeno significant change reported up to1990 chart (Rose), range drawn to its right edgesite optocoupler CTR records: 50 percent loss, median and 10 to 90 percentDisplacement damage sensitivity by device classFor each device class, a bar for the fluence range over which parts failed or changed by roughly 20 to 50 percent or more, a diamond at the lowest fluence at which a change was reported, and an outlined bar for the 1990 chart. Silicon classes in 1 MeV neutron equivalent fluence in silicon; compound semiconductor classes in the particle reported.1e81e91e101e111e121e131e141e151e16Silicon: 1 MeV neutron equivalent in siliconSignal and rectifier diodes1990 chart (Rose, Physitron), row SIGNAL DIODES: from 5.4e13, drawn to the right edge of the original chart at 1e15Failure or large change: forward voltage doubles at 1e13 in power rectifiers; switching diode forward voltage changes at 1e14 to 5e14 (fluence above 10 keV). [6]*Lowest fluence with a reported change: 1e11, reverse current rises in rectifiers rated above 10 A (fluence above 10 keV, as compiled). [6]*Zener and reference diodes1990 chart (Rose, Physitron), row REFERENCE DIODES: from 4.3e12, drawn to the right edge of the original chart at 1e15Failure or large change: reverse current up 10 times at 2e14; Zener voltage drifts beyond about 1e14 to 1e15. [6]*Lowest fluence with a reported change: 1e14, Zener voltage still within 1 to 2 percent (fluence above 10 keV). [6]*Low-fT bipolar transistors1990 chart (Rose, Physitron), row LO FT TRANSISTORS: from 9.8e10, drawn to the right edge of the original chart at 1e15Failure or large change: 50 percent gain loss at 4e12 to 5e12 in vertical NPN and PNP transistors and 5e12 to 7e12 in 2N2222A and 2N2907A (derived), about 1.6e12 extrapolated for a 4N49 phototransistor; a lateral PNP loses 20 to 33 percent at 5e11, more at low emitter-base bias. [7], [8], [9]Lowest fluence with a reported change: 3.1e10, about 2 percent gain loss in the phototransistor of a 4N49 (fT about 300 MHz); 10 percent at 4.4e11 to 7.4e11 in dielectrically isolated vertical transistors. [7], [8]High-fT bipolar transistors (microwave)1990 chart (Rose, Physitron), row HI FT TRANSISTORS: from 9.8e12, drawn to the right edge of the original chart at 1e15Failure or large change: about 40 percent gain loss in an IC NPN transistor near 8e13 (800 MeV protons). [10]*Lowest fluence with a reported change: 1e13, microwave transistors show no gain degradation below about 1e13 neutrons (as cited in the JPL chapter). [2]Bipolar power transistors1990 chart (Rose, Physitron), row POWER TRANSISTORS (BIPOLAR): from 9.8e10, drawn to the right edge of the original chart at 1e15Failure or large change: 50 percent gain loss in individual units from 8e10 to about 1e13; type medians 1.3e11 (2N3772) to 7e12 (2N3740). [11]Lowest fluence with a reported change: 9e9, 10 percent gain loss at 9e9 to 1.6e10 in 2N3772 (fT about 1.4 MHz; fast burst reactor fluence). [11]SiGe HBTsFailure or large change: 20 to 88 percent gain loss at 9.9e13 depending on the generation; minimal for the 0.12 um 8HP. [12]Lowest fluence with a reported change: 1.4e13, 5 to 28 percent gain loss (63 MeV protons; part of it ionization at the emitter-base spacer). [12]IGBTsFailure or large change: on-state voltage up 2.2 times and saturation current halved at 1e13. [13]Lowest fluence with a reported change: 1e12, saturation current down 15 to 20 percent. [13]JFETs1990 chart (Rose, Physitron), row JFET TRANSISTORS: from 9.8e13, drawn to the right edge of the original chart at 1e15Lowest fluence with a reported change: 6.5e12, gate leakage up about 1000 times in detector-grade JFETs (27 MeV protons, authors' 1 MeV equivalent). [14]No significant change reported up to 1e14: transconductance unchanged to 2.9e13; minimal gain change reported above 1e14. [10]*, [14]Si power MOSFETs, 150 to 500 V1990 chart (Rose, Physitron), row MOSFET TRANSISTORS: from 9.8e13, drawn to the right edge of the original chart at 1e15Failure or large change: on-resistance up 2 to 24 times at 5.4e14 in 150 to 250 V parts. [16]Lowest fluence with a reported change: 8e10, subthreshold leakage changes in an IRF640 (63 MeV neutrons); on-resistance up 10 to 30 percent at 1.1e13 to 3.3e13 in 400 to 500 V parts. [15], [16]Si power MOSFETs, 30 to 100 VFailure or large change: on-resistance up 1.2 to 4.7 times at 2.8e15. [16]Lowest fluence with a reported change: 1e15, on-resistance up about 20 percent. [16]Linear bipolar ICs1990 chart (Rose, Physitron), row LINEAR ICs (Si) (BIPOLAR): from 5.1e11, drawn to the right edge of the original chart at 1e15Failure or large change: LM137 catastrophic failure at 2.4e11 to 4.9e11; RH1056 at 3e11 to 1.2e12; LM124 input current up 132 nA at 1.7e12; narrow-base processes unaffected to 2.2e12. [17], [18]Lowest fluence with a reported change: 1.1e11, LM137 start-up and LM117 output shifts at 5e10 protons (50 MeV). [17]Digital bipolar ICs1990 chart (Rose, Physitron), row DIGITAL ICs (Si) (BIPOLAR): from 5.1e12, drawn to the right edge of the original chart at 1e15Failure or large change: I2L from 3e13, TTL 6e13 to 2e15, ECL 2.6e14 to 6e15, oxide-isolated logic 6e14 to 6e15 (failure ranges). [4], [5]Lowest fluence with a reported change: 6e12, failures of low-power TTL from about 6e12. [4], [5]CMOS ICs1990 chart (Rose, Physitron), row ALL MOS (Fig. 10-14): from 1.1e14, drawn to the right edge of the original chart at 1e15Failure or large change: n-MOS 1.1e14 to 1e15, bulk CMOS 2.4e14 to 2.2e15, CMOS/SOS 1.3e15 to 1e16 (failure ranges). [4], [5]Lowest fluence with a reported change: 8e10, SDRAM retention time falls from 3 s to 0.2 to 0.6 s (63 MeV neutrons). [15]Si photodiodes and phototransistorsFailure or large change: phototransistor gain down 40 to 60 percent at 5e11 and 60 to 80 percent at 1e12, at low photocurrent. [19]Lowest fluence with a reported change: 1.9e10, 4N49 phototransistor photoresponse down 10 percent (200 MeV protons). [7]CCDs (charge transfer)Failure or large change: significant charge transfer losses below 1 krad(Si) of protons, about 9e9; inefficiency 2.4e-5 at 1.2e10. [20]*, [21]*Lowest fluence with a reported change: 5e8, parallel charge transfer inefficiency up 1e-6 (derived from the damage constant of a science CCD, 10 MeV protons). [20]*CMOS image sensors (pinned photodiode)Failure or large change: quantum efficiency and full well degraded at 6.2e14. [22]Lowest fluence with a reported change: 6.2e9, mean dark current and random telegraph signal rise at the lowest fluence tested. [22]No significant change reported up to 1.9e12: quantum efficiency, conversion gain and full well unchanged to 1.9e12. [22]Optocouplers, amphoteric GaAs:Si LEDFailure or large change: 50 percent CTR loss at 1e10 to 1.8e10 protons (50 MeV) and 1.7e10 to 2.8e10 (63 MeV, low LED current). [7], [23], [24]Lowest fluence with a reported change: 4.9e9, 10 percent CTR loss at about 2.5e9 protons (63 MeV). [23]Optocouplers, AlGaAs double heterojunction LEDFailure or large change: 50 percent loss at 1.6e11 to 3.4e11 protons (50 MeV) and 4.7e11 to 6e11 neutrons. [24], [25]Lowest fluence with a reported change: 4.4e10, 6 to 15 percent loss at the lowest level tested (50 MeV protons); 10 percent at 8.5e10 to 1.3e11 neutrons. [24], [25]Optocouplers, GaAsP LEDFailure or large change: 50 percent loss at 3.3e12 to 3.4e12 neutrons, 90 percent at 6.5e12. [25]Lowest fluence with a reported change: 3.1e10, 3.5 percent loss in a 6N140 (155 MeV protons); 15 percent at 1.9e11. [7]Optocouplers, all LED types: site records1990 chart (Rose, Physitron), row OPTICAL ISOLATORS: from 2.7e10, drawn to the right edge of the original chart at 1e15Site records, 50 percent CTR loss: 5.3e11 at the median, 2.9e10 to 5.3e12 from the 90th to the 10th percentile part type (19 retention statements, 14 public). [33]10 percent CTR loss for the most sensitive decile of part types: 3.2e9. [33]Unijunction transistors (1990 chart only)1990 chart (Rose, Physitron), row UNIJUNCTION TRANSISTORS: from 5.8e11, drawn to the right edge of the original chart at 1e15SCRs (1990 chart only)1990 chart (Rose, Physitron), row SILICON-CONTROLLED RECTIFIERS: from 9.3e10, drawn to the right edge of the original chart at 1e15Compound semiconductors: particle as reportedLEDs, amphoteric GaAs:Si (50 MeV p)Failure or large change: 50 percent light loss, depending on bias, lot and current. [7], [26]Lowest fluence with a reported change: 2e9, about 10 percent light loss. [26]LEDs, AlGaAs double heterojunction (50 MeV p)Failure or large change: 50 percent light loss (50 to 58 MeV) for typical units. [26], [27]Lowest fluence with a reported change: 5e9, 2 to 3 percent light loss. [26]LEDs, GaAsP (50 MeV p)Lowest fluence with a reported change: 1.5e11, change visible; stated to be more than 10 times harder than amphoteric LEDs. [26]Laser diodes (50 MeV p)Failure or large change: threshold current up 20 to 30 percent, within a factor of 2 across six laser types. [28]Lowest fluence with a reported change: 3e11, threshold current up 2 to 7 percent. [28]GaAs ICs and MMICs (neutrons)1990 chart (Rose, Physitron), row GaAs ICs: from 9.3e13, drawn to the right edge of the original chart at 1e15Failure or large change: JFET transconductance down 20 percent from about 5e13 in lightly doped channels. [2]Lowest fluence with a reported change: 4e13, FET pinch-off voltage starts to fall. [2]No significant change reported up to 1e15: FET small-signal gain holds until near 1e15. [2]GaN HEMTs (neutrons; 10 MeV p)Failure or large change: 20 to 25 percent drain current and transconductance loss at 1e14 protons (10 MeV). [30]*Lowest fluence with a reported change: 6e12, drain current changes at 6e12 neutrons (1 MeV equivalent, compiled onset). [29]SiC power JFET (1 MeV n eq)Failure or large change: on-resistance 2.5 times at 6.6e13, 14.7 times at 1.7e14, non-functional above 4e14. [31]Lowest fluence with a reported change: 1.3e13, on-resistance up 30 percent (authors' 1 MeV equivalent). [31]InP HBTs (neutrons, 1 MeV eq)Failure or large change: gain below 20 (from about 200) at 2e14 to 6e14. [32]Lowest fluence with a reported change: 2.3e13, gain already halved at the lowest exposure. [32]1e81e91e101e111e121e131e141e151e16failure or large change (roughly 20 to 50 percent or more)lowest fluence with a reported changeno significant change reported up to1990 chart (Rose), range drawn to its right edgesite optocoupler CTR records: 50 percent loss, median and 10 to 90 percent
Silicon rows use 1 MeV neutron equivalent fluence in silicon: proton fluences converted with the silicon NIEL of Jun and co-workers (the tables of this page), reactor fluences as the source gave them (the 1971 handbook values for diodes are the fluence above 10 keV, not converted). Optocouplers combine a III-V LED with a silicon detector; they are converted with silicon NIEL, as the optocoupler CTR prior on the total dose priors page does, a conversion the optocoupler compendium of LaBel and co-workers calls high risk. Compound semiconductor rows are in the particle and energy reported, because NIEL scaling across materials and between particles in them is not reliable enough to place them on the silicon scale. The 1990 bars end at the right edge of the original chart, 1e15. On a desktop, hovering over a mark shows its values and sources; the table below lists them all.
The values behind each bar, and where they come from

Fluences in n/cm2 (1 MeV neutron equivalent in silicon) unless the class says otherwise. The 1990 values were read from the figures of the JPL chapter at 300 dpi (Fig. 10-13, and Fig. 10-14 for the IC families), to about 0.05 decade. The gold values are the optocoupler CTR prior of the total dose priors page: CTR/CTR0 = 1/(1 + K phi) fitted to retention statements from REDW, RADECS and GSFC test reports and from internal reports, which are anonymized; 50 percent loss is at phi = 1/K.

ClassLowest fluence with a reported changeFailure or large change1990 chart
Signal and rectifier diodes1e11: reverse current rises in rectifiers rated above 10 A (fluence above 10 keV, as compiled) [6]*1e13 to 5e14: forward voltage doubles at 1e13 in power rectifiers; switching diode forward voltage changes at 1e14 to 5e14 (fluence above 10 keV) [6]*from 5.4e13 (row as printed: SIGNAL DIODES)
Zener and reference diodes1e14: Zener voltage still within 1 to 2 percent (fluence above 10 keV) [6]*2e14 to 1e15: reverse current up 10 times at 2e14; Zener voltage drifts beyond about 1e14 to 1e15 [6]*from 4.3e12 (row as printed: REFERENCE DIODES). The 1990 chart also has temperature-compensated reference diodes from 6.6e11.
Low-fT bipolar transistors3.1e10: about 2 percent gain loss in the phototransistor of a 4N49 (fT about 300 MHz); 10 percent at 4.4e11 to 7.4e11 in dielectrically isolated vertical transistors [7], [8]1.6e12 to 7e12: 50 percent gain loss at 4e12 to 5e12 in vertical NPN and PNP transistors and 5e12 to 7e12 in 2N2222A and 2N2907A (derived), about 1.6e12 extrapolated for a 4N49 phototransistor; a lateral PNP loses 20 to 33 percent at 5e11, more at low emitter-base bias [7], [8], [9]from 9.8e10 (row as printed: LO FT TRANSISTORS)
High-fT bipolar transistors (microwave)1e13: microwave transistors show no gain degradation below about 1e13 neutrons (as cited in the JPL chapter) [2]8e13: about 40 percent gain loss in an IC NPN transistor near 8e13 (800 MeV protons) [10]*from 9.8e12 (row as printed: HI FT TRANSISTORS)
Bipolar power transistors9e9: 10 percent gain loss at 9e9 to 1.6e10 in 2N3772 (fT about 1.4 MHz; fast burst reactor fluence) [11]8e10 to 1e13: 50 percent gain loss in individual units from 8e10 to about 1e13; type medians 1.3e11 (2N3772) to 7e12 (2N3740) [11]from 9.8e10 (row as printed: POWER TRANSISTORS (BIPOLAR))
SiGe HBTs1.4e13: 5 to 28 percent gain loss (63 MeV protons; part of it ionization at the emitter-base spacer) [12]9.9e13: 20 to 88 percent gain loss at 9.9e13 depending on the generation; minimal for the 0.12 um 8HP [12]
IGBTs1e12: saturation current down 15 to 20 percent [13]1e13: on-state voltage up 2.2 times and saturation current halved at 1e13 [13]
JFETs6.5e12: gate leakage up about 1000 times in detector-grade JFETs (27 MeV protons, authors' 1 MeV equivalent) [14]; no significant change to 1e14: transconductance unchanged to 2.9e13; minimal gain change reported above 1e14 [10]*, [14]from 9.8e13 (row as printed: JFET TRANSISTORS)
Si power MOSFETs, 150 to 500 V8e10: subthreshold leakage changes in an IRF640 (63 MeV neutrons); on-resistance up 10 to 30 percent at 1.1e13 to 3.3e13 in 400 to 500 V parts [15], [16]3.3e13 to 5.4e14: on-resistance up 2 to 24 times at 5.4e14 in 150 to 250 V parts [16]from 9.8e13 (row as printed: MOSFET TRANSISTORS). The 1990 row is for MOSFETs in general, not power parts.
Si power MOSFETs, 30 to 100 V1e15: on-resistance up about 20 percent [16]2.8e15: on-resistance up 1.2 to 4.7 times at 2.8e15 [16]
Linear bipolar ICs1.1e11: LM137 start-up and LM117 output shifts at 5e10 protons (50 MeV) [17]2.4e11 to 1.7e12: LM137 catastrophic failure at 2.4e11 to 4.9e11; RH1056 at 3e11 to 1.2e12; LM124 input current up 132 nA at 1.7e12; narrow-base processes unaffected to 2.2e12 [17], [18]from 5.1e11 (row as printed: LINEAR ICs (Si) (BIPOLAR))
Digital bipolar ICs6e12: failures of low-power TTL from about 6e12 [4], [5]3e13 to 6e15: I2L from 3e13, TTL 6e13 to 2e15, ECL 2.6e14 to 6e15, oxide-isolated logic 6e14 to 6e15 (failure ranges) [4], [5]from 5.1e12 (row as printed: DIGITAL ICs (Si) (BIPOLAR)). The 1990 IC-family chart (Fig. 10-14) puts bipolar logic families between 4e12 (I2L) and 3e14 (ECL).
CMOS ICs8e10: SDRAM retention time falls from 3 s to 0.2 to 0.6 s (63 MeV neutrons) [15]1.1e14 to 2.2e15: n-MOS 1.1e14 to 1e15, bulk CMOS 2.4e14 to 2.2e15, CMOS/SOS 1.3e15 to 1e16 (failure ranges) [4], [5]from 1.1e14 (row as printed: ALL MOS (Fig. 10-14))
Si photodiodes and phototransistors1.9e10: 4N49 phototransistor photoresponse down 10 percent (200 MeV protons) [7]5e11 to 1e12: phototransistor gain down 40 to 60 percent at 5e11 and 60 to 80 percent at 1e12, at low photocurrent [19]
CCDs (charge transfer)5e8: parallel charge transfer inefficiency up 1e-6 (derived from the damage constant of a science CCD, 10 MeV protons) [20]*9e9 to 1.2e10: significant charge transfer losses below 1 krad(Si) of protons, about 9e9; inefficiency 2.4e-5 at 1.2e10 [20]*, [21]*For a science imager any measurable loss matters, so the onset is the practical limit.
CMOS image sensors (pinned photodiode)6.2e9: mean dark current and random telegraph signal rise at the lowest fluence tested [22]; no significant change to 1.9e12: quantum efficiency, conversion gain and full well unchanged to 1.9e12 [22]6.2e14: quantum efficiency and full well degraded at 6.2e14 [22]
Optocouplers, amphoteric GaAs:Si LED4.9e9: 10 percent CTR loss at about 2.5e9 protons (63 MeV) [23]2.2e10 to 5.5e10: 50 percent CTR loss at 1e10 to 1.8e10 protons (50 MeV) and 1.7e10 to 2.8e10 (63 MeV, low LED current) [7], [23], [24]
Optocouplers, AlGaAs double heterojunction LED4.4e10: 6 to 15 percent loss at the lowest level tested (50 MeV protons); 10 percent at 8.5e10 to 1.3e11 neutrons [24], [25]3.5e11 to 7.5e11: 50 percent loss at 1.6e11 to 3.4e11 protons (50 MeV) and 4.7e11 to 6e11 neutrons [24], [25]
Optocouplers, GaAsP LED3.1e10: 3.5 percent loss in a 6N140 (155 MeV protons); 15 percent at 1.9e11 [7]3.3e12 to 6.5e12: 50 percent loss at 3.3e12 to 3.4e12 neutrons, 90 percent at 6.5e12 [25]
Optocouplers, all LED types: site records3.2e9: 10 percent CTR loss for the most sensitive decile of part types [33]2.9e10 to 5.3e12, median 5.3e11: 50 percent CTR loss from the most to the least sensitive decile of part types (19 retention statements, 14 from public reports) [33]from 2.7e10 (row as printed: OPTICAL ISOLATORS)
Unijunction transistors (1990 chart only)from 5.8e11 (row as printed: UNIJUNCTION TRANSISTORS)
SCRs (1990 chart only)from 9.3e10 (row as printed: SILICON-CONTROLLED RECTIFIERS)
LEDs, amphoteric GaAs:Si (50 MeV p)2e9: about 10 percent light loss [26]8e9 to 2.6e10: 50 percent light loss, depending on bias, lot and current [7], [26]
LEDs, AlGaAs double heterojunction (50 MeV p)5e9: 2 to 3 percent light loss [26]3e11 to 6e11: 50 percent light loss (50 to 58 MeV) for typical units [26], [27]
LEDs, GaAsP (50 MeV p)1.5e11: change visible; stated to be more than 10 times harder than amphoteric LEDs [26]
Laser diodes (50 MeV p)3e11: threshold current up 2 to 7 percent [28]3e12 to 1e13: threshold current up 20 to 30 percent, within a factor of 2 across six laser types [28]
GaAs ICs and MMICs (neutrons)4e13: FET pinch-off voltage starts to fall [2]; no significant change to 1e15: FET small-signal gain holds until near 1e15 [2]5e13: JFET transconductance down 20 percent from about 5e13 in lightly doped channels [2]from 9.3e13 (row as printed: GaAs ICs)
GaN HEMTs (neutrons; 10 MeV p)6e12: drain current changes at 6e12 neutrons (1 MeV equivalent, compiled onset) [29]1e14: 20 to 25 percent drain current and transconductance loss at 1e14 protons (10 MeV) [30]*
SiC power JFET (1 MeV n eq)1.3e13: on-resistance up 30 percent (authors' 1 MeV equivalent) [31]6.6e13 to 4e14: on-resistance 2.5 times at 6.6e13, 14.7 times at 1.7e14, non-functional above 4e14 [31]
InP HBTs (neutrons, 1 MeV eq)2.3e13: gain already halved at the lowest exposure [32]2.3e13 to 6e14: gain below 20 (from about 200) at 2e14 to 6e14 [32]

Values marked * came from sources read through a web summary while this chart was built and are to be checked against the full text. No public chart or table of displacement damage thresholds by technology published after 2000 was found; the bar charts before Rose 1990 (Long 1980 [5] and the "traditional bar chart" of Raymond and Petersen 1987 [4]) are, like it, ranges of failure fluences, not onsets.

  1. J. R. Srour, C. J. Marshall and P. W. Marshall, "Review of Displacement Damage Effects in Silicon Devices," IEEE Trans. Nucl. Sci. 50(3), p. 653, 2003. The Messenger and Spratt gain relation is its Eq. 3, p. 660, after G. C. Messenger and J. P. Spratt, Proc. IRE 46, 1958.
  2. C. Barnes and L. Selva, "Radiation Effects in MMIC Devices," Chapter 10 of the GaAs MMIC Reliability Assurance Guideline for Space Applications, JPL Publication 96-25, p. 203, 1996 (Figs. 10-13 and 10-14, p. 219 and 220; GaAs FET and JFET data, p. 233 and 234).
  3. M. Rose, "Updated Bar Charts of Device Radiation Thresholds," Physitron Corp., San Diego, CA, 1990, as cited in the JPL chapter above; the report itself was not found, and no description of how its bars were compiled was found.
  4. J. P. Raymond and E. L. Petersen, "Comparison of Neutron, Proton and Gamma Ray Effects in Semiconductor Devices," IEEE Trans. Nucl. Sci. 34(6), p. 1622, 1987.
  5. D. M. Long, "State-of-the-Art Review: Hardness of MOS and Bipolar Integrated Circuits," IEEE Trans. Nucl. Sci. 27(6), p. 1674, 1980.
  6. C. L. Hanks and D. J. Hamman, "Radiation Effects Design Handbook, Section 1. Semiconductor Diodes," NASA CR-1785, 1971. Read through a web summary for this chart; the values marked * are to be checked against the full text.
  7. B. G. Rax, C. I. Lee, A. H. Johnston and C. E. Barnes, "Total Dose and Proton Damage in Optocouplers," IEEE Trans. Nucl. Sci. 43(6), p. 3167, 1996.
  8. C. N. Arutt, K. M. Warren, R. D. Schrimpf, R. A. Weller, J. S. Kauppila, J. D. Rowe, A. L. Sternberg, R. A. Reed, D. R. Ball and D. M. Fleetwood, "Proton Irradiation as a Screen for Displacement-Damage Sensitivity in Bipolar Junction Transistors," IEEE Trans. Nucl. Sci. 62(6), p. 2498, 2015.
  9. H. J. Barnaby, R. D. Schrimpf, K. F. Galloway, X. Li, J. Yang and C. Liu, "Displacement Damage in Bipolar Junction Transistors: Beyond Messenger-Spratt," IEEE Trans. Nucl. Sci. 64(1), p. 149, 2017.
  10. H. Spieler, "Introduction to Radiation-Resistant Semiconductor Devices and Circuits," Lawrence Berkeley National Laboratory tutorial, about 1997. Read through a web summary for this chart; the values marked * are to be checked against the full text.
  11. C. Rosenberg, I. Arimura and A. M. Unwin, "Statistical Analysis of Neutron-Induced Gain Degradation of Power Transistors," IEEE Trans. Nucl. Sci. 17(6), p. 160, 1970.
  12. Y. Lu, J. D. Cressler, R. Krithivasan, Y. Li, R. A. Reed, P. W. Marshall, C. Polar, G. Freeman and D. Ahlgren, "Proton Tolerance of Third-Generation, 0.12 um 185 GHz SiGe HBTs," IEEE Trans. Nucl. Sci. 50(6), p. 1811, 2003.
  13. A. R. Hefner, D. L. Blackburn and K. F. Galloway, "The Effect of Neutrons on the Characteristics of the Insulated Gate Bipolar Transistor (IGBT)," IEEE Trans. Nucl. Sci. 33(6), p. 1428, 1986.
  14. G. F. Dalla Betta, M. Manghisoni, L. Ratti, V. Re, V. Speziali, G. Traversi and A. Candelori, "Proton-Induced Damage in JFET Transistors and Charge Preamplifiers on High-Resistivity Silicon," IEEE Trans. Nucl. Sci. 51(5), p. 2880, 2004.
  15. J.-P. David, F. Bezerra, E. Lorfevre, R. Durand and T. Dargnies, "Experimental Review of Si Commercial Devices Damage Sensitivity," RADECS 2005 Proceedings, paper PB2, 2005.
  16. J. E. Gillberg, D. I. Burton, J. L. Titus, N. Hubbard and C. F. Wheatley, "Updated Responses of Devices from the FSG and FSP Radiation-Hardened Power MOSFET Families to 1-MeV Equivalent Neutrons," IEEE Radiation Effects Data Workshop, p. 138, 2002.
  17. B. G. Rax, A. H. Johnston and T. Miyahira, "Displacement Damage in Bipolar Linear Integrated Circuits," IEEE Trans. Nucl. Sci. 46(6), p. 1660, 1999.
  18. H. J. Barnaby, R. D. Schrimpf, A. L. Sternberg, V. Berthe, C. R. Cirba and R. L. Pease, "Proton Radiation Response Mechanisms in Bipolar Analog Circuits," IEEE Trans. Nucl. Sci. 48(6), p. 2074, 2001.
  19. F. Irom, G. R. Allen, L. D. Edmonds and B. G. Rax, "Proton Damage in LED and Phototransistor of Micropac 66179 Optocoupler," IEEE Radiation Effects Data Workshop, 2016.
  20. P. Verhoeve, T. Prod'homme, T. Oosterbroek, N. Boudin and L. Duvet, "CCD Characterization for Astronomy Space Missions at ESA," Proc. SPIE 9154, 2014. Read through a web summary for this chart; the values marked * are to be checked against the full text.
  21. C. J. Marshall and P. W. Marshall, "Proton Effects and Test Issues for Satellite Designers, Part B: Displacement Effects," IEEE NSREC Short Course, Section IV, 1999. Read through a web summary for this chart; the values marked * are to be checked against the full text.
  22. C. Virmontois, V. Goiffon, F. Corbiere, P. Magnan, S. Girard and A. Bardoux, "Displacement Damage Effects in Pinned Photodiode CMOS Image Sensors," IEEE Trans. Nucl. Sci. 59(6), p. 2872, 2012.
  23. K. A. LaBel, S. D. Kniffin, R. A. Reed, H. S. Kim, J. L. Wert, D. L. Oberg, E. Normand, A. H. Johnston, G. K. Lum, R. Koga, et al., "A Compendium of Recent Optocoupler Radiation Test Data," IEEE Radiation Effects Data Workshop, p. 123, 2000.
  24. T. F. Miyahira and A. H. Johnston, "Trends in Optocoupler Radiation Degradation," IEEE Trans. Nucl. Sci. 49(6), p. 2868, 2002.
  25. M. Poizat, M. Sauvagnac, A. Samaras, Y. Padie, P. Garcia, B. Renaud, L. Gouyet, J. P. Abadi, et al., "Compendium of Total Ionizing Dose, Displacement Damage and Single Event Transient Test Data of Various Optocouplers for ESA," IEEE Radiation Effects Data Workshop, 2013.
  26. A. H. Johnston, B. G. Rax, L. E. Selva and C. E. Barnes, "Proton Degradation of Light-Emitting Diodes," IEEE Trans. Nucl. Sci. 46(6), p. 1781, 1999.
  27. H. N. Becker and A. H. Johnston, "Proton Damage in LEDs With Wavelengths Above the Silicon Wavelength Cutoff," IEEE Trans. Nucl. Sci. 51(6), p. 3558, 2004.
  28. A. H. Johnston, T. F. Miyahira and B. G. Rax, "Proton Damage in Advanced Laser Diodes," IEEE Trans. Nucl. Sci. 48(6), p. 1764, 2001.
  29. J. W. McClory and J. C. Petrosky, "Temperature Dependent Electrical Characteristics of Neutron Irradiated AlGaN/GaN HFETs," IEEE Trans. Nucl. Sci. 54(6), p. 1969, 2007.
  30. B. D. Weaver, T. J. Anderson, A. D. Koehler, J. D. Greenlee, J. K. Hite, D. I. Shahin, F. J. Kub and K. D. Hobart, "On the Radiation Tolerance of AlGaN/GaN HEMTs," ECS J. Solid State Sci. Technol. 5(7), p. Q208, 2016. Read through a web summary for this chart; the values marked * are to be checked against the full text.
  31. S. Popelka, P. Hazdra, R. Sharma, V. Zahlava and J. Vobecky, "Effect of Neutron Irradiation on High Voltage 4H-SiC Vertical JFET Characteristics: Characterization and Modeling," IEEE Trans. Nucl. Sci. 61(6), p. 3030, 2014.
  32. A. Shatalov, S. Subramanian and A. Dentai, "Neutron Irradiation Effects in InP/InGaAs Single Heterojunction Bipolar Transistors," IEEE Trans. Nucl. Sci. 47(6), p. 2551, 2000.
  33. Space RHA, "Total Dose Hardness Priors," optocoupler CTR prior (tid-priors-tool.html): CTR retention statements from REDW, RADECS and GSFC test reports and from internal test reports, the internal ones anonymized, fitted as CTR/CTR0 = 1/(1 + K phi).
How the numbers are computed

NIEL at an energy is interpolated log-log between the two tabulated energies that bracket it, in the table for the selected material. An energy below or above a material's table gives out of range; nothing is extrapolated.

A compound uses Bragg's rule: its NIEL is the sum over its elements of the mass fraction times the element's NIEL, with mass fractions niAi / Σ njAj from the stoichiometric coefficients you enter and the atomic masses. The compound is out of range wherever any of its elements is, and its curve in the figures is evaluated at the tabulated energies of its elements where all of them have data.

The damage energy cross section is NIEL × A / NA, shown in MeV-mb (1 mb = 10-27 cm²), with A the atomic mass of the element or the formula mass of the compound.

Fluence conversion: Φtarget = Φsource × NIEL(Esource) / NIEL(Etarget), the fluence at the target energy that does the same displacement damage in the same material. Displacement damage dose: DDD = Φ × NIEL(E), in MeV/g.

1 MeV neutron equivalent fluence: Φneq = Φp × NIELp(E) / NIEL1MeV n,Si, with NIEL(1 MeV neutron, Si) = 1.95 × 10-3 MeV-cm²/g. The 95 MeV-mb reference of ASTM E722 converts to 2.04 × 10-3 MeV-cm²/g; with this tool's value the equivalent fluence is about 4.5 percent higher (multiply by 0.957 to convert). The hardness factor is NIELp(E) / NIEL1MeV n,Si. The reference is the silicon value for every material, as the result label says.

Data and provenance

Proton NIEL values, in MeV-cm²/g, are from Jun et al. (2003) for ten elements: silicon, carbon, aluminum, phosphorus, gallium, germanium, arsenic, indium, copper and selenium. Every element carries every row of Table I of the paper (silicon 36 rows, 0.2 keV to 1000 MeV), transcribed from the published table and checked against it on 29 September 2026. Each element carries the displacement threshold energy Ed of its calculation, shown in the plot legend and the data table headings.

Disclaimer

This tool is provided for preliminary analysis and educational purposes only. NIEL data is from Jun et al. “Proton Nonionizing Energy Loss (NIEL) for Device Applications,” IEEE Trans. Nucl. Sci., vol. 50, no. 6, pp. 1924–1928, Dec. 2003. Users are solely responsible for verifying all outputs against their own analysis and applicable standards before making any design, test, or mission decisions. Space RHA LLC makes no warranties regarding accuracy or fitness for purpose. See the Help & Reference page for methodology details and references.

References

I. Jun, M. A. Xapsos, S. R. Messenger, E. A. Burke, R. J. Walters, G. P. Summers and T. Jordan, “Proton Nonionizing Energy Loss (NIEL) for Device Applications,” IEEE Trans. Nucl. Sci., vol. 50, no. 6, pp. 1924–1928, Dec. 2003.

ASTM E722, Standard Practice for Characterizing Neutron Fluence Spectra in Terms of an Equivalent Monoenergetic Neutron Fluence for Radiation-Hardness Testing of Electronics: the 95 MeV-mb reference for 1 MeV neutrons in silicon.

For preliminary analysis and education; verify every output against your own analysis before a design, test or mission decision. The help page covers NIEL physics, Bragg's rule, fluence conversion, DDD and the neutron equivalent.Open the help page