K. Ketata

Comparative analysis of accelerated ageing effects on power RF LDMOS reliability

We present in this paper results of comparative reliability study of three accelerated ageing tests applied on power RF LDMOS: Thermal Shock Tests (TST, air-air test), Thermal Cycling Tests (TCT, air-air test) and High Temperature Storage Life (HTSL). The two first tests are carried out with a drain current flowing through the device during stress. The results obtained show the variation and the Devices performance quantitative shifts for some macroscopic electric parameters such as threshold voltage (V th ), transconductance (G m ), drain-source current (I ds ), on-state resistance (R ds on) and feedback capacitance (C rs ) tinder various ageing tests. To understand the degradation phenomena that appear after ageing, we used a new electro-thermal model implemented in Agilents ADS as a reliability tool.

Study of RF N− LDMOS critical electrical parameter drifts after a thermal and electrical ageing in pulsed RF

An innovative reliability test bench dedicated to RF power devices is currently implemented. This bench allows to apply both electric and thermal stress for lifetime test under radar pulsed RF conditions. This paper presents the first investigation findings of critical electrical parameter degradations after thermal and electrical ageing. It shows that the tracking of a set of parameters (drain–source current, on-state resistance, threshold voltage, feedback capacitance and

Deep levels induced by low energy B+ implantation into Ge-preamorphised silicon in correlation with end of range formation

Abstract It is well established that low energy B+ ion implantation into Ge- (or Si) implantation pre-amorphised silicon allows ultra-shallow p+n junctions formation. However, this process is known to generate defects such as dislocation loops, vacancies and interstitials which can act as vehicles to different mechanisms inducing electrically active levels into the silicon bulk. The junctions studied have been obtained using 3 keV/ 10 15 cm −2 B + implantation into Ge-implantation pre-amorphised substrates and into a reference crystalline substrate. Accurate measurements using deep level transient spectroscopy (DLTS) and isothermal transient capacitance ΔC(t,T) were performed to characterise these levels. Such knowledge is crucial to improve the device characteristics. In order to sweep the silicon band gap, various experimental conditions were considered. The analysis of DLTS spectra have first showed three deep levels associated to secondary induced defects. Their concentration profiles were derived from isothermal transient capacitance at depths up to 3.5 μm into the silicon bulk and allowed us to detect a new deep level. The evolution of such defect distribution in correlation with the technological steps is discussed. The end of range (EOR) defect influence on electrical activity of secondary induced defects in ultra-shallow p+n diodes is clearly demonstrated.

Electrical parameters degradation of power RF LDMOS device after accelerated ageing tests

This paper reports novel methods for accelerated ageing tests, with comparative reliability between them for stresses applied on power RF LDMOS: Thermal Shock Tests (TST), Thermal Cycling Tests (TCT), High Voltage Drain (HVD) and coupling thermal and electrical effects under various conditions. The investigation findings obtained after various ageing tests show the degradation and the devices performance shifts for most important electric parameters such as transconductance (Gm), on-state resistance (R ds_on ), feedback capacitance (C rss ) and gate-drain capacitance (C gd ). This means that the tracking of these parameters enables to consider the hot carrier injection as the dominant degradation phenomenon. However, this is explained by excitation and trapping of electrons in the oxide-silicon interface at the drain side. A physical simulation software (2D, Silvaco-Atlas) has been used to locate and confirm degradation phenomena.

A model for diffusion of beryllium in InGaAs/InP heterostructures

Abstract This study reports on Be diffusion from a Be-doped (3×10 19 cm −3 ) In 0.53 Ga 0.47 As layer sandwiched between undoped InP layers grown by gas source molecular beam epitaxy. To explain the obtained experimental depth profiles, a kick-out model of substitutional interstitial diffusion mechanism, involving neutral Be interstitials for the InGaAs epilayer and singly positively charged Be interstitials for the InP epilayers, is proposed. Using the boundary conditions at the heterojunctions and taking into account the built-in electric field, Fermi level and bulk self-interstitial generation/annihilation effects, we obtained a good agreement between the simulated and experimental data.

Determination of beryllium and self-interstitial diffusion parameters in InGaAs

Abstract The diffusion of Be in InGaAs grown by gas source molecular beam epitaxy has been studied. The observed secondary ion mass spectrometry Be profiles, obtained for annealing cycles with a temperature range of 700–900°C, could be explained considering several forms of kick-out mechanism. The beryllium diffusion models in InGaAs have been obtained without a priori hypothesis on interstitial beryllium and self-interstitial. Moreover, the fitting procedure of experimental data has been described precisely. We conclude that two kick-out mechanisms lead to similar fittings of experimental profiles and may not be distinguishable using our experimental conditions.

Modeling of the transient enhanced diffusion of boron implanted into preamorphized silicon: the case of BF2+ implantation

We have simulated transient enhanced diffusion (TED) in the presence of end-of-range (EOR) defects produced by Ge amorphization followed by BF2 implantation. Ostwald ripening of EOR defects has been taken into account. A comparison of annealed profiles with equivalent B implantation shows that the existing models are not sufficient to simulate the BF2 experimental profiles where the boron diffusion depth is very low. We have proposed that the presence of fluorine can act as sinks for interstial boron and, hence, reduces the boron diffusion depth in order to obtain a good approximation of experimental profiles.

Be diffusion in InGaAs layers grown by gas source molecular beam epitaxy

The diffusion of Be from buried Be-doped InGaAs epitaxial layers, grown by gas source molecular beam epitaxy, has been studied for temperatures between 700 and 900°C. A kick-out model, involving neutral Be interstitial species and singly positively charged group III self-interstitials, is proposed for the diffusion mechanism.

Be diffusion in InGaAs, InGaAsP epitaxial layers and across InGaAs/InGaAsP, InGaAs/InP heterointerfaces

Abstract The diffusion of Be during post-growth Rapid Thermal Annealing (RTA) in InGaAs, InGaAsP, InGaAs/InGaAsP and InGaAs/InP epitaxial structures grown by Gas Source Molecular Beam Epitaxy (GSMBE) has been studied. The observed Secondary Ion Mass Spectrometry (SIMS) concentration distributions, obtained for annealing cycles with time durations of 10–240 s and temperatures in the range of 700–900°C for Be doping concentration of 3×10 19 cm −3 , could be explained by kick-out mechanism considering the neutral Be interstitial species and positively charged group-III self-interstitials.

Beryllium diffusion in InGaAs/InGaAsP structures grown by gas source molecular beam epitaxy

A systematic study of Be post-growth diffusion from buried Be-doped InGaAs layers in undoped InGaAsP layers grown by gas source molecular beam epitaxy was carried out. The experimental structures consisted of a 2000 A Be-doped (3×1019 cm−3) In0.53Ga0.47As layer sandwiched between 5000 A undoped In0.73Ga0.27As0.58P0.42 layers. The samples were subjected to rapid thermal annealing in the temperature range from 700 to 900°C with time durations of 10–240 s. Secondary ion mass spectrometry was employed for a quantitative determination of the Be depth profiles. To explain the obtained experimental results, the kick-out model of substitutional–interstitial diffusion mechanism, involving neutral Be interstitial species and positively charged Ga and In self-interstitial species, has been considered. The Be and self-interstitial diffusivities, the rate coefficient of self-interstitial generation or annihilation, the self-interstitial equilibrium concentration, and the intrinsic carrier concentration were obtained for ternary and quaternary layers as functions of temperature.