Joints and magnetics · Workhorse

Kovar (FeNiCo)

Kvvalues describe: ASTM F15

Vacuum electronics & X-ray sourcesQuantum hardwareVacuum equipment & components

Why it wins

CTE matched to borosilicate and alumina across the braze range: the glass-to-metal and ceramic-to-metal seal ring that survives thermal cycling.

Why not the alternative

A stainless ring on alumina cracks the ceramic on cooldown; copper is too soft to constrain.

Watch out

Ferromagnetic: keep away from beams and qubits; rusts unplated.

Properties

The values below are candidate: compiled from the sources named, not yet individually validated. Provisional provenance: ASTM F15 and Carpenter Kovar data; Electron tube and ceramic-to-metal sealing practice.

Wet cleaning

Recipe	alkaline degrease, DI rinse, then wet-hydrogen fire at 1000–1100 °C; controlled oxidation before glass sealing
Forbidden	leaving carbon on the surface before firing: it ruins the seal
Limit	the oxide layer is grown deliberately, so cleaning is a step in a sequence, not an end

Vacuum and outgassing

Outgassing, unbaked (10 h)	2e-9 mbar·L/s/cm²
Outgassing, baked	1e-11 mbar·L/s/cm²
Vapour pressure	negligible

Temperature

Bake, assembled	450 °C
Vacuum degas	900 °C
Braze / H2 firing	1100 °C
Metallurgical limit	Curie 435 °C

Thermal

CTE	5.5 ppm/K
Thermal conductivity	17 W/m·K
Specific heat	439 J/kg·K
Emissivity	0.2–0.3
Melting / softening	1450 °C

Mechanical

Strength	345 ys MPa
Tensile	517 MPa
Elongation	25 %
Young's modulus	138 GPa
Hardness	130–160 HV
Density	8.36 g/cm3

Electrical and magnetic

Relative permeability	ferromagnetic
Resistivity	49 µΩ·cm

Engineering

Corrosion	poor bare; always plated or sealed
Joining	glass sealing and brazing are its purpose
Process notes	oxidised deliberately before glass sealing
Availability and cost	stock as sealing alloy

Brazing

Sourced pairings

With	Filler	Atmosphere	Expansion gap at set
Alumina 94 to 99.8 percent	Cusil	vacuum or H2 (brazing), wet H2 (metallisation)	0.28 % at 780 °C Kovar (FeNiCo) on the outside undetermined
Alumina 94 to 99.8 percent	Ticusil	vacuum 1e-5 mm Hg or inert gas	0.28 % at 780 °C Kovar (FeNiCo) on the outside undetermined
Sapphire	Incusil-ABA	vacuum furnace, oxygen-free	0.04 % at 605 °C Kovar (FeNiCo) on the outside matched
Alumina 94 to 99.8 percent	Incusil 15	vacuum	0.10 % at 605 °C Kovar (FeNiCo) on the outside undetermined
Graphite, pyrolytic graphite, glassy carbon / CVD diamond windows / Silicon carbide (CVD) / Aluminum nitride and Shapal	Ticusil	vacuum 1e-5 mm Hg or inert gas	0.42 % at 780 °C Kovar (FeNiCo) on the outside over budget
Oxygen-free copper C10100 / C10200	Cusil or Nioro or Palcusil 15	vacuum, H2 or inert gas	0.62 % at 780 °C Oxygen-free copper C10100 / C10200 on the outside over budget
Austenitic stainless 304L / 316L / 316LN ESR	Nioro or Palcusil 15	vacuum, H2 or inert gas	0.74 % at 850 °C Austenitic stainless 304L / 316L / 316LN ESR on the outside over budget
Schott 8250	named only in prose: none, glass-to-metal seal	per the glass-sealing process	no filler identified, nothing to derive
Molybdenum and TZM one built object stud and hub assembly of a rotating X-ray tube anode, the helical preform sitting in the cavity of the Kovar nut	50Cu-40Pd-10Ni	vacuum	set temperature absent: no solidus recorded for cu-pd-ni-50-40-10
Alumina 94 to 99.8 percent one built object MC 2935 Kovar header assembly and its pin, The Bendix Corporation, Kansas City Division: a metallized alumina insulator brazed to a Kovar pin and header to form a vacuum-tight electrical feedthrough	Nicusil 3	dry hydrogen	set temperature absent: no solidus recorded for ag-cu-ni-nicusil-3
Alumina 94 to 99.8 percent one built object ASTM F19 tensile buttons of 94 percent alumina with Fe-29Ni-17Co interlayers, Sandia National Laboratories, brazed by three methods compared side by side: molybdenum-manganese metallization with nickel plate, thin-film PVD metallization, and active filler metal brazing	Cusil or 35Au/65Cu or Incusil-ABA or Cusil-ABA or 97Ag-1Cu-2Zr or 35Au-62Cu-2Ti-1Ni	dry hydrogen on the metallized and thin-film routes, vacuum or partial-pressure argon on the active route	0.28 % at 780 °C Kovar (FeNiCo) on the outside undetermined
Alumina 94 to 99.8 percent one built object ASTM F19 tensile buttons and flush-mount cylinder assemblies of alumina ceramic active-brazed to Kovar interlayers, Sandia National Laboratories, testing conformal ALD alumina coatings against braze filler metal run-out	97Ag-1Cu-2Zr	oil-free, top-loading, cryogenic-pumped high-vacuum furnace, typical absolute pressures 5e-7 to 1e-6 torr during the peak brazing soak, monitored on a NIST-traceable ionization gauge	set temperature absent: no solidus recorded for ag-cu-zr-97-1-2
Oxygen-free copper C10100 / C10200 one built object second joint of the same Jefferson Lab FEL waveguide window: the OFHC copper frame carrying the BeO disc, brazed onto the copper-plated Kovar flange that provides the vacuum seal	Incusil 10	vacuum braze oven	0.59 % at 685 °C Oxygen-free copper C10100 / C10200 on the outside over budget
Alumina 94 to 99.8 percent stated as a pairing	named only in prose: not named by the table. Table 15.1 varies the METALLIZING MIXTURE and its sintering temperature, and the braze that attaches the Kovar to the metallized specimen is not printed.	not stated by the source	no filler identified, nothing to derive

Preparation

Mo-Mn 80/20 fired 1500 C under wet H2, coating 0.025-0.038 mm, Ni plating 0.003 mm refired 1000 C
no metallisation
none
Mo-Mn then Ni plating
not stated by the manufacturer
not stated
pre-oxidation of the Kovar
washer preform between stud and hub, helical preform in the cavity of the nut
molybdenum-manganese metallization fired in wet hydrogen, then nickel plating, electrolytic from a nickel-chloride solution or electroless nickel-boron, the two compared with no difference in tensile strength attributable to either, brazing cycle of 10 minutes at 55 C above the liquidus, the parts placed into a preheated furnace and cooled in a water-cooled hydrogen-purged chamber
for the metallized route, a coating of molybdenum, sometimes tungsten, and manganese particles mixed with silicates and glass additives, fired in wet hydrogen at 1450 to 1600 C to leave a glassy metallic coating 300 to 500 micro-inches thick, then plated with 0.001 to 0.003 inch of nickel sinter-fired at 850 to 950 C in hydrogen. For the thin-film route, a PVD titanium layer 0.05 to 0.25 micrometres thick under a noble over-layer of gold, platinum or palladium 0.25 to 1.0 micrometres thick.
alumina cleaned in a three-step solvent process then resintered either in wet hydrogen at a 28 C dew point, 1500 C for 60 minutes, or air-fired at 1575 C for 120 minutes, Kovar pickled in deionized water and hydrochloric acid then bright-dipped in an acetic, nitric and hydrochloric solution, joint preload about 30 g per square centimetre, part of the Kovar carries a conformal ALD alumina coating of 1 to 20 nm. Cycle: 15 C/min to 925 C with a 15 minute soak, 10 C/min to 985 C with a 5 minute soak, 25 C/min back to 925 C, then furnace cool.
mask the copper plated KOVAR and apply a light blasting with aluminum oxide particles to clean the beryllia surface
seven metallizing mixtures formulated at the Sperry Gyroscope Company, sintered between 1300 and 1600 C. Composition 65, 292.5 g Mo and 7.5 g Ti at 1500 C. Composition 91, 270 g Mo and 30 g LiMnO3 at 1500 C. Composition 141, 291 g Mo and 9 g talc MgO-SiO2 at 1600 C. Composition 72, 240 g Mo and 73.6 g CeO2 at 1500 C. Composition 50, 255 g Mo and 48 g SiO2 and 22 g Mn, at 1300 C on one body and 1500 C on another. Composition 49, 255 g Mo and 48 g SiO2 and 26 g MnO at 1500 C.

Compatibility

What documents state about this material faced with a filler metal, without the other member of the joint. A statement that names the exact filler comes first. A statement about a whole AWS class comes after it: a class covers dozens of alloys, so it is the weaker of the two, and the corpus never turns one into the other.

listed35Au/65Cu
Kohl lists this filler for copper, Kovar and nickel brazes.
Conditions and source
Book page 192, table XVI, Brazing filler metals for high-vacuum systems. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer, so it relays a manufacturer selection list and not a measurement by the author. The Applications column names base metals and never names the other member of the joint. Row 8 of the table, composition Au 35 and Cu 65, which matches the corpus designation 35Au/65Cu (BAu-3) by name. The corpus entry is statut no-data, motif no source read this pass, so this compatibility statement is the only thing the corpus carries about that alloy and nothing here supplies its melting range.
[A]W. H. Kohl, Soldering and brazing, Vacuum, volume 14, number 5, pages 175 to 198, 1964, Pergamon Press, PII 0042-207X(64)90858-9
Peer-reviewed review article on soldering and brazing for vacuum devices, republished later in the Handbook of Vacuum Physics edited by A. H. Beck. Table XVI, Brazing filler metals for high-vacuum systems, book page 192, lists 21 filler metals with liquidus, solidus and an Applications column. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer. It is therefore a manufacturer selection list relayed by the author, not a measurement of his own, and the Applications column names base metals without naming a partner. The page of each statement is carried by the entry that uses it, never by this registry line.
listedBAu-1
Kohl lists this filler for copper, Kovar and nickel.
Conditions and source
Book page 192, table XVI, Brazing filler metals for high-vacuum systems. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer, so it relays a manufacturer selection list and not a measurement by the author. The Applications column names base metals and never names the other member of the joint. Row 9 of the table, composition Au 37.5 and Cu 62.5, which matches the corpus entry BAu-1 exactly.
[A]W. H. Kohl, Soldering and brazing, Vacuum, volume 14, number 5, pages 175 to 198, 1964, Pergamon Press, PII 0042-207X(64)90858-9
Peer-reviewed review article on soldering and brazing for vacuum devices, republished later in the Handbook of Vacuum Physics edited by A. H. Beck. Table XVI, Brazing filler metals for high-vacuum systems, book page 192, lists 21 filler metals with liquidus, solidus and an Applications column. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer. It is therefore a manufacturer selection list relayed by the author, not a measurement of his own, and the Applications column names base metals without naming a partner. The page of each statement is carried by the entry that uses it, never by this registry line.
wetsNioro
Kohl states that this filler will wet tungsten and molybdenum as well as copper, Kovar, nickel and stainless steel, with excellent flow.
Conditions and source
Book page 192, table XVI, Brazing filler metals for high-vacuum systems. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer, so it relays a manufacturer selection list and not a measurement by the author. The Applications column names base metals and never names the other member of the joint. Row 13 of the table, composition Ni 18 and Au 82, which matches the corpus entry Nioro BAu-4 exactly.
[A]W. H. Kohl, Soldering and brazing, Vacuum, volume 14, number 5, pages 175 to 198, 1964, Pergamon Press, PII 0042-207X(64)90858-9
Peer-reviewed review article on soldering and brazing for vacuum devices, republished later in the Handbook of Vacuum Physics edited by A. H. Beck. Table XVI, Brazing filler metals for high-vacuum systems, book page 192, lists 21 filler metals with liquidus, solidus and an Applications column. The table carries its own reserve, printed under it: adapted from the list of a commercial manufacturer. It is therefore a manufacturer selection list relayed by the author, not a measurement of his own, and the Applications column names base metals without naming a partner. The page of each statement is carried by the entry that uses it, never by this registry line.

Chemical affinity

What documents state about a BINARY SYSTEM, one element of a filler metal faced with one element of this material. An affinity is not a verdict on a joint: a joint brings many binaries together at once, some helpful and some harmful, and what follows is the list of those the corpus can name. The count below is not written anywhere. It falls out of the crossing of this material's elements with the composition of every filler the corpus holds, so it grows on its own the day a binary enters the corpus.

Elements of this material: Fe Ni Co Mn Si, read from a composition table.

dissolution and erosion21 fillersCu against Ni, Cu against Fe1 source+2 statements

The molten filler dissolves the base, or the base dissolves into the filler far enough to change what the filler is. The joint loses base metal, or the filler stops flowing.

A copper filler takes nickel into solution faster than it takes iron. It picks up enough nickel to raise its own liquidus and stops flowing before it has gone far.Cu against Ni

21 fillers: 10Sn-30Cu-60Ag, 13In-27Cu-60Ag, 35Au-62Cu-2Ti-1Ni, 35Au/65Cu, 49Ti-49Cu-2Be, 50Cu-40Pd-10Ni, 97Ag-1Cu-2Zr, Ag-31.5Cu-10Pd, BAu-1, BCu-1, Cusil, Cusil-ABA, Cusin-1 ABA, Incusil 10, Incusil 15, Incusil-ABA, Nicoro-80, Nicusil 3, Palcusil 10, Palcusil 15, Ticusil

Conditions and verbatim

Book page 87, section on nickel and high-nickel alloys. The chapter states that the copper filler metal characteristically alloys to a greater extent with nickel than with iron, that alloying during brazing makes capillary flow difficult, and that the copper does not flow far before it has picked up enough nickel to raise its liquidus and reduce its fluidity. The chapter treats this as a workable process with a design consequence rather than a prohibition: place the filler as close to the joint as possible, keep a sufficient reservoir, and heat as rapidly as practicable. Page 67 names the same couple among its examples of interaction and adds that the remelt temperature of the filler-metal layer ends up higher than its original solidus.

Source[A] M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

Source conditionsChapter 4 of the second edition, book pages 63 to 162 in the copy read, which treats base metals family by family and states metallurgical reactions, atmosphere limits and filler selection guides. It is a handbook chapter, not a qualification record. The page of each statement is carried by the entry that uses it, never by this registry line.

An iron base metal dissolves in molten copper. On a graphite joint the dissolved iron then grows a columnar iron-copper-carbon phase at the graphite interface, which is what makes the joint possible where pure copper alone does not wet.Cu against Fe

Conditions and verbatim

Book page 158, in the passage that opens a new brazing technique for joining graphite to itself or to metals such as molybdenum, tungsten or copper. The chapter states that it is essentially impossible to braze graphite with copper filler metal AWS BCu-1 because no wetting occurs, and that HOWEVER, when a graphite base material is combined with an iron base metal in copper brazing, the iron base metal dissolves in molten copper, the dissolved iron growing as part of a columnar Fe6-9 Cu-1.6C alloy phase at the graphite interface at a constant brazing temperature. The corpus already carries the first half of that passage as the does-not-wet entry of graphite against BCu-1, whose conditions noted that the iron route existed and was NOT recorded. This entry records the mechanism, and the two are the two halves of one paragraph.

Source[A] M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

brittle intermetallic11 fillersTi against Fe, Al against Fe1 source+2 statements

The two elements combine into a compound that is harder and less tolerant of strain than either of them. It sits in the filler, in the base next to the interface, or in both.

Titanium and the iron of a stainless steel form intermetallic compounds at the joint. They grow with time at temperature, and they form in the base metal next to the interface as well as in the filler.Ti against Fe

9 fillers: 35Au-62Cu-2Ti-1Ni, 48Ti-48Zr-4Be, 49Ti-49Cu-2Be, 56Zr-28V-16Ti, 71.5Ti-28.5Ni, Cusil-ABA, Cusin-1 ABA, Incusil-ABA, Ticusil

Conditions and verbatim

Book page 147, conclusions of a study on titanium brazed to 304 stainless steel, citing Ref 226. The chapter states that an increase in copper concentration resulted in an increase in titanium content in the titanium over 304 stainless steel filler metal and enhanced the formation of intermetallic compounds between titanium and 304 stainless steel, and that to maximize shear strength the brazing time must be limited to a maximum of 15 minutes. It states separately that intermetallic compounds are not limited to the filler metal and can also form in the base metal next to the interface. The study joins a titanium MEMBER to a stainless MEMBER with a silver-copper filler, so the titanium is on the base side there. This corpus keys the entry on the element pair, which is what the named compounds are made of, and the direction of supply is not part of the fact.

Source[A] M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

Aluminium and the iron of a stainless steel grow an iron-aluminium intermetallic layer at the interface. Once that layer becomes continuous the joint is significantly fragile, and it takes about ten minutes at 600 C to get there.Al against Fe

2 fillers: 4047, BAlSi-2

Conditions and verbatim

Book page 147. The chapter states that strong joints between stainless steel and aluminum via furnace brazing can be produced using a eutectic aluminum-silicon filler metal, provided brazing times are kept sufficiently short to avoid formation of the second, more fragile iron-aluminum intermetallic layer, and that at a brazing temperature of 600 C brazing times should remain less than approximately 10 minutes. The temperature carried is the one the time limit is stated AT, not a threshold above which the mechanism starts. The chapter adds on the preceding page that the growth of this second layer is controlled by diffusion through it once it has formed, and that the shear strength peaks at 21 MPa after a 10 minute hold.

Characteristic temperature600 °C, M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

Source[A] M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

grain-boundary penetration1 fillerB against Fe1 source+1 statement

An element of the filler runs along the grain boundaries of the base as a low-melting liquid, ahead of any bulk dissolution. Thin sections are the ones it destroys.

Boron in a filler penetrates the grain boundaries of a ferrous base as a low-melting complex and degrades the joint. A thin base metal is where the damage is worst.B against Fe

1 filler: BNi-2

Conditions and verbatim

Book page 67, in the section on base-metal and filler-metal interactions. The chapter states that in brazing of ferrous-base high-temperature alloys with filler metals containing boron, grain-boundary penetration of the base metal by a low-melting complex can cause joint degradation, and that this effect is particularly damaging if the base metal is thin, as in the case of brazed honeycomb sandwich panels. The chapter names ferrous-base HIGH-TEMPERATURE alloys, not every ferrous base. The corpus keys the entry on iron because that is the element the sentence names, and this condition records that the document had a narrower family in view.

Source[A] M. M. Schwartz, Brazing, 2nd edition, ASM International, 2003, chapter 4, Base Metals and Base-Metal Family Groups, DOI 10.1361/brse2003p063

Open in the brazing abacus →

Grades

ASTM F15 (1.3981)	the sealing alloy spec: control anneal and decarburization for glassing

Sources · 1

O Hanlon, A Users Guide to Vacuum Technology

Used by

Tools

Brazing AbacusKovar
Brazing Route EngineKovar

Essays

← The Matter