698 Electronics Handbook
degradation effects the parts should not be periodically handled or tested; only electrically test devices prior
to use.
Component failures, such as pin to pin leakage, can also be caused by contaminating films that are on
the seal and embedment glass of the package. One such failure mode reported for SRAMs was caused
by lead sulfide from antistatic ABS/PVC (acrylonitrile-butadiene-styrene/polyvinyl chloride) plastic trays
used to store parts during assembly and postseal operations. The lead sulfide exhibited itself as a shiny film
over the seal and embedment glass and resulted in more than 20 µA (at 5 V) leakage. PVC plastic that uses
Thioglycolate (or Dibutyltin bis or Iso Octylthioglicolate, heat stabilizers), a common additive, can cause
sulfur-based contamination (and corrosion) on parts with lead-based embedment glass. Because of this,
PVC manufacturers are pursuing alternate stabilizer materials, and products that can have performance
degradations are often stored and processed using metal trays.
7.10.6 Lead Finish
Reducing lead in the environment is an environmental health and safety issue that is of prime importance
today. While a typical microprocessor has approximately 0.2 g of lead and a computer motherboard has
2 to 3 g of lead, and the whole USA electronic interconnection market uses less than 2% of the world’s
lead production, to address environmental concerns, there are various environmental directives limiting
the use of lead and other hazardous substances. These directives, formulated by the European Union in a
Directive of the European parliament and of the Council of the European Union are:
r End of Life Vehicles (ELV) Directive 2000/53/EC in force since October 21, 2000, requires that
products be free of heavy metals such as mercury, cadmium, hexavalent chromium and lead. It
requires a recycling system be established by vehicle manufacturers for vehicles made after July 1,
2002 and for all vehicles regardless of date of manufacture by July 1, 2007. Lead can still be used as
an alloying additive in copper and in solderable applications.
r The WEEE Directive, Waste Electrical and Electronic Equipment (WEEE) Directive 2002/96/EC,
expands the recycling requirements of the ELV Directive to include a broad range of electronic
and electrical appliances and equipment. WEEE went into effect on February 13, 2003. It is to be
scheduled to become European national law by August 13, 2004, and be applicable to consumer
use products by August 13, 2005. Article 2(3) however states “Equipment which is connected with
the protection of the essential interests of the security of Member States, arms, munitions and war
material shall be excluded from this Directive. This does not, however, apply to products which are
not intended for specifically military purposes.”
r RoHS Directive, The Restriction of Hazardous Substances in Electrical and Electronic Equipment,
this Directive 2002/95/EC establishes standards and limits for the hazardous material content in
electronic and electrical equipment. The Directive went into effect on February 13, 2003. It is
scheduled to become European national law by August 13, 2004 for be in force for products by July
1, 2006. Banned or restricted substances include lead, mercury, cadmium, hexavalent chromium,
certain brominated flame retardants (PBBs), and polybrominated diphenyl ethers (PBDEs).
The recommended lead-free solder formulation is Sn-Ag-Cu for board assembly but there are other
formulations such as Nickel-Palladium (NiPd), or Nickel-Palladium with Gold flash (NiPdAu). Passive
components, to be compatible with a lower temperature Lead process (which is 215◦ C for 50/50 Tin/Lead
formulations and 230◦ C for 40/60 formulations) and the higher lead-free process of up to 260◦ C, use pure
matte Tin for their contacts. The use of lead in solder is partially based on several potential reliability issues.
Pure Tin component leads have been shown to result in inter-metallic migration in the termination of the
electronic component and the growth of tin whiskers which could cause short circuits (which is why there
is a exemption for military use (only) components).
The National Electronics Manufacturing Initiative (NEMI) has addressed the problem of “tin whiskers”
in lead-free assemblies. A tin whisker is defined by them as
Copyright 2005 by Taylor & Francis Group
, Semiconductor Devices and Circuits 699
A spontaneous columnar or cylindrical filament, which rarely branches, of monocrystalline tin emanat-
ing from the surface of a plating finish. Furthermore, tin whiskers may have the following characteristics:
r an aspect ratio (length/width) greater than 2
r can be kinked, bent, twisted
r generally have consistent cross-sectional shape
r may have striations or rings around it
Their recommended test method is
r temperature cycling (−55◦ C to + 85◦ C, approximately 3 cycles/hour)
r temperature humidity tests at 60◦ C/93% RH
r ambient storage (air-conditioned facility)
(Note: Tin will undergo a phase transformation, becoming a powdery form (called Tin Pest) if Tin plated
parts are stored for more than 1 week at temperatures below 13◦ C).
7.10.7 Screening and Rescreening Tests
In the 1970s, the Navy instituted a policy requiring devices be rescreened on receipt (by the contractor or
an independent test laboratory) because there was evidence that Joint Army Navy (JAN) qualified devices
(including JANTX, JANTXV discrete component) were not of a quality level to be used in military hardware
(DoD directive 4245.7-M Transition from Development to Production, which was signed in 1984 and saw
wide implementation by the end of 1986).
Rescreening tests often imposed included the following:
Destructive physical analysis (DPA) examination tests, where two pieces of each lot of parts (as a
minimum) were cut open and examined to determine the workmanship of the parts. If the workmanship
was judged to be poor, the whole lot of parts were rejected (a lot of parts is defined as parts from one
manufacturer from one assembly line and one date code).
Particle impact noise detection (PIND) tests, where each hybrid part or IC with a cavity where the die
was unglassivated (insulated with a glass coating, Fig. 7.160 shows an IC with cracks in this coating), was
vibrated and transducers, mounted on the part would detect if there were any particles rattling around.
Parts with loose pieces in them were rejected from the shipment.
Go-no go electrical tests, and static and dynamic tests were required to be performed at low, ambient
(25◦ C) and high temperatures.
Hermeticity testing was required to test the integrity of the package seal.
To review the reasons for component rescreening we need to examine test data for the time period
involved. In 1981 it was reported that the Naval Weapons Support Center (NWSC) in Crane, Indiana,
found part defect rates of 15% for ICs and 17% for discrete semiconductors. In a January 1983 report,
one independent test laboratory found reject rates of 16.7% for linear ICs, 8.4% for digital ICs, and 9.2%
for CMOS ICs. By October 1984, this same laboratory found defect rates of 5.5% for linear ICs, 3.7%
for digital ICs, 8.2% for transistors, and 3.8% for optoelectronics. In 1983, the Defense Electronic Center
found semiconductor defect rates of 8% for nonmilitary specification parts, with a less than 1% defect rate
for military specification devices. In 1986, a military contractor estimated rescreening failures of 0.9% for
ICs and 1.5% for transistors. In 1989, a manufacturer of a special purpose computer workstation for the
U.S. Navy reported the following rescreening results: IC/semiconductors tested: 8312 (127 types of parts);
total rejects PIND/DPA: 25 (16 PIND failures on hybrid components, 6 electrical failures, 3 mechanical
failures—broken leads); reject rate: 0.30%.
Breaking down the hybrid failures, of 99 military qualified oscillators tested, 6 were rejected for a rejec-
tion rate of 6.1%; 53 nonmilitary qualified oscillators were tested of which 10 were rejected, for a rejection
rate of 18.8%.
The Semiconductor Industry of America conducted a quality statistics program to monitor and report
on industry data on various quality control indices and parameters of microcircuits from 1985 to the early
Copyright 2005 by Taylor & Francis Group
degradation effects the parts should not be periodically handled or tested; only electrically test devices prior
to use.
Component failures, such as pin to pin leakage, can also be caused by contaminating films that are on
the seal and embedment glass of the package. One such failure mode reported for SRAMs was caused
by lead sulfide from antistatic ABS/PVC (acrylonitrile-butadiene-styrene/polyvinyl chloride) plastic trays
used to store parts during assembly and postseal operations. The lead sulfide exhibited itself as a shiny film
over the seal and embedment glass and resulted in more than 20 µA (at 5 V) leakage. PVC plastic that uses
Thioglycolate (or Dibutyltin bis or Iso Octylthioglicolate, heat stabilizers), a common additive, can cause
sulfur-based contamination (and corrosion) on parts with lead-based embedment glass. Because of this,
PVC manufacturers are pursuing alternate stabilizer materials, and products that can have performance
degradations are often stored and processed using metal trays.
7.10.6 Lead Finish
Reducing lead in the environment is an environmental health and safety issue that is of prime importance
today. While a typical microprocessor has approximately 0.2 g of lead and a computer motherboard has
2 to 3 g of lead, and the whole USA electronic interconnection market uses less than 2% of the world’s
lead production, to address environmental concerns, there are various environmental directives limiting
the use of lead and other hazardous substances. These directives, formulated by the European Union in a
Directive of the European parliament and of the Council of the European Union are:
r End of Life Vehicles (ELV) Directive 2000/53/EC in force since October 21, 2000, requires that
products be free of heavy metals such as mercury, cadmium, hexavalent chromium and lead. It
requires a recycling system be established by vehicle manufacturers for vehicles made after July 1,
2002 and for all vehicles regardless of date of manufacture by July 1, 2007. Lead can still be used as
an alloying additive in copper and in solderable applications.
r The WEEE Directive, Waste Electrical and Electronic Equipment (WEEE) Directive 2002/96/EC,
expands the recycling requirements of the ELV Directive to include a broad range of electronic
and electrical appliances and equipment. WEEE went into effect on February 13, 2003. It is to be
scheduled to become European national law by August 13, 2004, and be applicable to consumer
use products by August 13, 2005. Article 2(3) however states “Equipment which is connected with
the protection of the essential interests of the security of Member States, arms, munitions and war
material shall be excluded from this Directive. This does not, however, apply to products which are
not intended for specifically military purposes.”
r RoHS Directive, The Restriction of Hazardous Substances in Electrical and Electronic Equipment,
this Directive 2002/95/EC establishes standards and limits for the hazardous material content in
electronic and electrical equipment. The Directive went into effect on February 13, 2003. It is
scheduled to become European national law by August 13, 2004 for be in force for products by July
1, 2006. Banned or restricted substances include lead, mercury, cadmium, hexavalent chromium,
certain brominated flame retardants (PBBs), and polybrominated diphenyl ethers (PBDEs).
The recommended lead-free solder formulation is Sn-Ag-Cu for board assembly but there are other
formulations such as Nickel-Palladium (NiPd), or Nickel-Palladium with Gold flash (NiPdAu). Passive
components, to be compatible with a lower temperature Lead process (which is 215◦ C for 50/50 Tin/Lead
formulations and 230◦ C for 40/60 formulations) and the higher lead-free process of up to 260◦ C, use pure
matte Tin for their contacts. The use of lead in solder is partially based on several potential reliability issues.
Pure Tin component leads have been shown to result in inter-metallic migration in the termination of the
electronic component and the growth of tin whiskers which could cause short circuits (which is why there
is a exemption for military use (only) components).
The National Electronics Manufacturing Initiative (NEMI) has addressed the problem of “tin whiskers”
in lead-free assemblies. A tin whisker is defined by them as
Copyright 2005 by Taylor & Francis Group
, Semiconductor Devices and Circuits 699
A spontaneous columnar or cylindrical filament, which rarely branches, of monocrystalline tin emanat-
ing from the surface of a plating finish. Furthermore, tin whiskers may have the following characteristics:
r an aspect ratio (length/width) greater than 2
r can be kinked, bent, twisted
r generally have consistent cross-sectional shape
r may have striations or rings around it
Their recommended test method is
r temperature cycling (−55◦ C to + 85◦ C, approximately 3 cycles/hour)
r temperature humidity tests at 60◦ C/93% RH
r ambient storage (air-conditioned facility)
(Note: Tin will undergo a phase transformation, becoming a powdery form (called Tin Pest) if Tin plated
parts are stored for more than 1 week at temperatures below 13◦ C).
7.10.7 Screening and Rescreening Tests
In the 1970s, the Navy instituted a policy requiring devices be rescreened on receipt (by the contractor or
an independent test laboratory) because there was evidence that Joint Army Navy (JAN) qualified devices
(including JANTX, JANTXV discrete component) were not of a quality level to be used in military hardware
(DoD directive 4245.7-M Transition from Development to Production, which was signed in 1984 and saw
wide implementation by the end of 1986).
Rescreening tests often imposed included the following:
Destructive physical analysis (DPA) examination tests, where two pieces of each lot of parts (as a
minimum) were cut open and examined to determine the workmanship of the parts. If the workmanship
was judged to be poor, the whole lot of parts were rejected (a lot of parts is defined as parts from one
manufacturer from one assembly line and one date code).
Particle impact noise detection (PIND) tests, where each hybrid part or IC with a cavity where the die
was unglassivated (insulated with a glass coating, Fig. 7.160 shows an IC with cracks in this coating), was
vibrated and transducers, mounted on the part would detect if there were any particles rattling around.
Parts with loose pieces in them were rejected from the shipment.
Go-no go electrical tests, and static and dynamic tests were required to be performed at low, ambient
(25◦ C) and high temperatures.
Hermeticity testing was required to test the integrity of the package seal.
To review the reasons for component rescreening we need to examine test data for the time period
involved. In 1981 it was reported that the Naval Weapons Support Center (NWSC) in Crane, Indiana,
found part defect rates of 15% for ICs and 17% for discrete semiconductors. In a January 1983 report,
one independent test laboratory found reject rates of 16.7% for linear ICs, 8.4% for digital ICs, and 9.2%
for CMOS ICs. By October 1984, this same laboratory found defect rates of 5.5% for linear ICs, 3.7%
for digital ICs, 8.2% for transistors, and 3.8% for optoelectronics. In 1983, the Defense Electronic Center
found semiconductor defect rates of 8% for nonmilitary specification parts, with a less than 1% defect rate
for military specification devices. In 1986, a military contractor estimated rescreening failures of 0.9% for
ICs and 1.5% for transistors. In 1989, a manufacturer of a special purpose computer workstation for the
U.S. Navy reported the following rescreening results: IC/semiconductors tested: 8312 (127 types of parts);
total rejects PIND/DPA: 25 (16 PIND failures on hybrid components, 6 electrical failures, 3 mechanical
failures—broken leads); reject rate: 0.30%.
Breaking down the hybrid failures, of 99 military qualified oscillators tested, 6 were rejected for a rejec-
tion rate of 6.1%; 53 nonmilitary qualified oscillators were tested of which 10 were rejected, for a rejection
rate of 18.8%.
The Semiconductor Industry of America conducted a quality statistics program to monitor and report
on industry data on various quality control indices and parameters of microcircuits from 1985 to the early
Copyright 2005 by Taylor & Francis Group
