Oshida Y. Bioscience and Bioengineering of Titanium Materials

Подождите немного. Документ загружается.

[9-107] Oshida Y, Farzin-Nia F. Progressive damage assessment of TiNi

endodontic files. In: Shape memory implants. Yahia L. editor. New York:

Springer. 2000. pp. 236249.

[9-108] Bensmann G, Baumgart F. Osetosyntheseklammern aus Nicekltitan

Herstellung, Vorversuche und klinischer Einsatz. Tech Mitt Krupp

1982;40:123134.

[9-109] Barouk LS. The double compressive nickel-titanium shape-memory sta-

ple in foot surgery. In: Shape memory implants. Yahia L. editor. New

York: Springer. 2000. pp. 162173.

[9-110] Lévesque J, Dubé D, Fiset M, Mantovani D. Materials and properties for

coronary stents. Adv Mater Process 2004;162:4548.

Other Applications 281

Else_BBTM-OSHIDA_ch009.qxd 10/5/2006 4:04 PM Page 281

Chapter 10

Fabrication Technologies

10.1. Casting 285

10.2. Machining 288

10.3. Electro Discharge Machining (EDM) and CAD/CAM 289

10.4. Isothermal Forging 291

10.5. Superplastic Forming (SPF) 291

10.6. Diffusion Bonding (DB) 294

10.7. Powder Metallurgy 295

10.8. Metal Injection Molding (MIM) 297

10.9. Laser Welding/Forming 299

10.10. Soldering 301

10.11. Heat Treatment (HT) 302

References 303

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 283

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 284

285

Chapter 10

Fabrication Technologies

There are many conventional and advanced technologies available, including cast-

ing, milling, spark erosion, forging, powder metallurgy, a combination of milling

or spark erosion with laser welding [10-1–10-6] fabricating medical, and dental

prostheses and/or devices. This chapter will discuss and review these technologies.

10.1. CASTING

There are unusual features associated with titanium casting. Unlike castings of

other metals, (i) there are no commercial titanium alloys developed strictly for

casting applications, therefore all titanium castings have compositions based on

those of the common wrought alloys, and (ii) titanium castings are equal or nearly

equal in strength to their wrought counterparts. The casting of titanium dental

appliances was noted in the early 1970s with the works of Waterstratt [10-7],

followed by numerous studies in Japan, Europe, and United States toward the pre-

cision casting of dental prostheses, and the development of casting machines and

suitable investment materials. There are a number of ways to cast titanium. At this

moment, there are about 10 different systems available for casting titanium [10-8,

10-9]. Among them, for example, the Ohara (atmospheric argon shielded arc melt-

ing followed by centrifugal casting) and the Castmatic (pressurized argon shielded

arc melting followed by gravity casting accelerated by two atmospheres argon

pressure above and one atmosphere vacuum below) are examples of such

machines. These casting machines are designed and manufactured in particular to

overcome the challenging parameters such as high activity and low density of tita-

nium. It was found that the dental titanium castings in these machines are suscep-

tible to surface oxygen contamination and possible interaction with the

investment, causing a tripling of surface hardness (about 600 Knoop hardness

number), whereas the rest of the casing (200 KHN) at 200 m underneath the sur-

face case [10-10, 10-11].

Evaluation of titanium casting via radiographic means is a technique that has

been found useful in detecting internal porosities [10-12]. Sprue design in Ti, as

in gold casting is important in the prevention of porosities. The most favorable

sprue design was determined to be a rounded or curved sprue [10-13]. While many

authors discuss porosity as a challenge when casting Ti, one study found that 97%

of the removable partial denture (RPD) framework cast was technically acceptable

[10-14]. Cecconi et al. [10-14] evaluated the degree of porosity via radiographic

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 285

examination of 300 cast CpTi (grade 2) RPD frameworks. A single-chamber, gas

pressure Titec 205 M casting machine was used. One lab technician cast all

frameworks. The frameworks were used clinically. It was reported that (i) no

framework was retuned for any reason, (ii) 250 out of 300 frameworks were tech-

nically acceptable as cast, and (iii) 41 were technically acceptable after laser weld-

ing modifications and nine were technically unacceptable after casting and needed

to be redone.

The castability of titanium and its alloys can be metallurgically improved

[10-15, 10-16]. Since commercially pure titanium has a relatively high melting

point (1670°C), meta-stable -phase Ti-based alloys have recently been developed

to reduce the melting points by adding Cr or Pd (e.g., Ti-20Cr-0.2Si, or Ti-23Pd-

5Cr) [10-17, 10-18]. It was reported that Ti casting was done by the argon-arc

melting and subsequently argon or vacuum-pressurized casting of Ti-6Al-4V, Ti-

15V, Ti-20Cu, and Ti-30Pd [10-19], and NiTi, Ti-Pt, Ti-Cr, and Ti-Zr alloys [10-

20], as well as CpTi [10-21, 10-22]. However, special care must always be taken

to control casting defects with titanium (which can not be totally avoided when

using a casting process). Defects include such things as shrinkage cavities, pin-

holes, or voids, and they are particularly a problem with denture bases [10-23,

10-24], where the surface area is relatively large compared to the thin thickness.

It is important to avoid the creation of an alpha layer (or -case), eliminate poros-

ity, and achieve a good fit of the casting onto a die. Since titanium is an active

metal, particularly with oxygen, the investment casting of Ti is adversely affected

by reaction of the melt with the atmosphere in the melting or casting chamber,

reaction of the melt with the melting crucible, and reaction of the melt with the

mold materials [10-2, 10-25–10-29]. Even when refractories having thermody-

namic stability similar to that of the titanium oxide family are used (such as MgO

and Al

), the very strong reducing power of titanium decomposes such common

oxides. The oxygen liberated from the oxide diffuses into the surface of the cast-

ing to form a hardened layer (about 200 m thick), depending on the investment

used [10-10, 10-11, 10-30].

Such an oxygen-rich layer (-case) possesses several drawbacks, including (i)

unnecessary hard and brittleness [10-31], (ii) reducing ductility and fatigue resist-

ance [10-9, 10-32], and (iii) resulting in inferior titanium/porcelain bond strengths

[10-33–10-35]. Several methods have been evaluated to minimize the formation of

this reaction layer, including coating the wax pattern with an oxide that is thermo-

dynamically more stable than the titanium oxides before investing [10-36]. When

a wax pattern was coated with a slurry containing Y

powder, it was reported

that (i) the surface hardness (25 m from the cast surface) of cast commercially

pure titanium was found to be remarkably reduced (280 VHN) compared to the

surface hardness of the uncoated cast Ti (530 VHN) [10-11], and (ii) use of ZrO

286 Bioscience and Bioengineering of Titanium Materials

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 286

reduced the hardness near the cast surface (457 VHN) [10-37]. Miyakawa et al.

[10-29] investigated reaction layer structures of titanium cast in three molds con-

taining spinel (MgO·Al

). For this spinel mold, the reaction zone consisted of

baked product layer and the alpha case, which was divided into two layers. The

outside was stabilized and hardened by a solid solution of Al, whereas the inside

was not so much hardened without a solid solution of Al. It was found that, for the

investment mixed with magnesium acetate solution, the Al-stabilized alpha case

with higher hardness value formed on limited surfaces [10-29]. A novel invest-

ment material was developed by Nakamura et al. [10-38]. Fine particle sizes of

alumina (Al

) and zirconia (ZrO

) were imbedded among relatively large parti-

cle size silica (SiO

) crystals, so that the direct contact between molten titanium

and silica can be avoided to a great extent. Zr metal powder was mixed with mag-

netia (magnesium oxide: MgO) to provide an expansion-controlling mold material

[10-38]. It was also reported that, using a large particle size of colloidal silica as

a mixing liquid to phosphate-bonded investment, thermal expansion can be con-

trolled and surface roughness, as well as the surface reaction layer, was improved

[10-39]. Cast titanium for dental crowns has been investigated using the invest-

ment (consisted essentially of SiO

, Mg

, SiO

·H

O and Mg

SiO

), claiming

that formation of -case is reduced [10-40]. A relatively simple method with an

investment casting of titanium was developed with various characteristics (such as

arc melting in argon, combination of vacuum and pressured assisted casting, and

shell casting with zirconia), and a three-unit crown was successfully cast. It was

claimed that the formation of a brittle -case was avoided [10-41].

There are several unique melting methods. Traditionally, titanium alloys are

melted, refined, and cast into round ingots following two or more runs of the vac-

uum arc remelting process. Multiple vacuum arc remelting runs are used to

homogenize the chemical compositions and, further, in an attempt to overcome

two types of melt defects – type I hard  inclusions and high-density inclusions –

that are a primary quality concern when producing high-performance titanium

alloys. These melt defects, when present undetected at sub-surface location, could

promote premature fatigue failure in rotating components, shorten the life cycle of

a component, and compromise product performance and safety. Sampath [10-42]

introduced a competing melting method as plasma arc cold hearth melting and

electron beam cold hearth melting, followed by a vacuum arc remelting. Melting

Ti-Nb, Ti-Ta, or Ti-Zr binary alloy is not easy because of melting point and den-

sity differences between Ti and these alloying elements are large. To overcome this

technical problem, a cold crucible levitation melting, using cast molding made of

a mixture of alumina cement and calcium zirconate with a lithium carbonate as a

liquid was developed, and successful melting without any segregation was

reported [10-43].

Fabrication Technologies 287

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 287

10.2. MACHINING

Conventionally, research and development of dental materials has been focused on

casting, with very few activities in machinability, specifically aiming on

CAD/CAM materials. For an ordinary machining process, one should remember

that titanium’s work hardening rate (n = 0.05 and n is in the equation  ⫽ K

) is

much less than that of 316 (18Cr-8Ni-2Mo) stainless steel (n ⫽ 0.45), copper

(n ⫽ 0.54), and alpha brasses (n ⫽ 0.49), but with proper technique it can readily

be machined or polished to a very fine finish. At the same time, it should be

noticed that Ti is a poor conductor of heat (0.16 cal/cm/s·°Ccm

vs. 0.71 for

gold). While this allows excellent localized electric arc spot welding, it means that

during cutting heat will not dissipate quickly [10-10]. It is also true that the rela-

tively low value of modulus of elasticity (MOE) of titanium causes a tendency for

work to move away from the cutting tool. But if one uses a few basic rules - low

cutting speeds, high feed rates, copious amounts of cutting fluid, and sharp tools

replaced at the first signs of wear – good results are obtainable [10-10]. In general,

poor machinability can arise for various reasons, including high tensile strength,

high high-temperature strength, high hardness, large ductility, low value of modu-

lus of elasticity, high work-hardenability, low thermal conductivity, large thermal

expansion coefficient, and large reactivity with tool materials. Among these rea-

sons, low value of MOE, low thermal conductivity and high reactivity are typified

for titanium materials. Titanium materials have unique machining properties.

While the cutting forces are only slightly higher than in machining steels, there are

other characteristics that make these alloys more difficult to machine than steels

having an equivalent hardness. For example, the chip-tool contact area while turn-

ing a titanium alloy is only about one-third to one-half as great as that of turning

steel. Also, the thermal conductivity of titanium alloy (21.9 W/(m·K)) is about

one-fourth of that of steels (80.4 W/(m·K)). This combination of a small contact

area and the low thermal conductivity result in very high cutting temperatures. At

a cutting speed of 100 ft/min, the temperature developed at the cutting edge of a

carbide tool is 537°C when cutting a steel, while on the titanium alloy, the tem-

perature can reach 704°C. Hence, the cutting speeds on titanium alloys must be

lower in order to maintain a tool–chip temperature below that which results in

short tool life [10-44]. Normally, S, Pb, P, Bi, Te, or rare-earth elements are alloyed

to improve the machinability. For Ti materials, it was reported that alloying with

Au, Ag, or Cu (as -eutectoid type) or with Nb or Hf (as  solid solution type)

improved the machinability [10-45].

For the final polishing of titanium materials as well as titanium casts, chemical

polishing and buff-polishing are normally practiced. As a novel polishing method, an

alcohol-based electrolyte was used while stirring; the original average roughness of

288 Bioscience and Bioengineering of Titanium Materials

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 288

CpTi (0.38 m) was refined to 0.03 m, and the 0.55 m original roughness of Ti-

6Al-4V was improved to be 0.12 m [10-46]. To prevent the implant periodontitis,

due to plaque accumulation, a certain portion of dental implants need to be finished

as smooth as possible to reduce the surface energy. Kyo et al. [10-47] developed the

electrolytic in-process dressing (ELID) grinding technique to the mirror surface pol-

ishing in the neck region of the NiTi alloy in contact with alveolar bone. It was

reported that (i) the ten-point average surface roughness was 185 nm and the center

line surface roughness was 22 nm, indicating that good surface properties can be

obtained, (ii) the results of analysis of the ELID ground surface using AES showed

oxygen peaks other than the titanium and nickel elements, and (iii) when compared

to the ground surface, the Ti peak was found to change only slightly along the depth,

while the oxygen peak decreased and the Ni peak increased, indicating that the

ground surface is oxidized during the ELID machining [10-47].

Since all materials have a finite modulus of elasticity, plastic deformation due to

machining is followed by some elastic recovery when the load is removed. In bend-

ing, the recovery is called springback. Springback in forming titanium and titanium

alloys increases as the R/T ratio (R represents the bend radius and T the work-metal

thickness) and yield strength increases and the elastic modulus decreases, and

inversely with forming temperature [10-48]. Springback in titanium alloys is more

difficult to predict than springback in steel, although it depends on the same prin-

ciples. Differences in the yield strength of various heats in titanium can cause dif-

ferences in springback; higher stress ratios of yield strength to tensile strength

generally result in greater springback. The springback phenomenon is mostly

important when orthodontic arch wires are concerned. Orthodontists usually acti-

vate orthodontic wires somewhat into the permanent deformation range.

Consequently, the practical working range is considered to be the elastic strain at

the yield strength of the wire, namely the quotient of yield strength (

) and MOE.

This important property (

/MOE) is termed springback [10-48].

10.3. ELECTRO DISCHARGE MACHINING (EDM) AND CAD/CAM

EDM, also known as spark erosion dentistry, has been used in industry for over 40

years as a means of machining hard metals with electrical impulses used to cut

metal passively and precisely into various shapes and forms [10-49]. In 1982,

Rubeling [10-50] introduced spark erosion and its application to the dental tech-

nology. Spark erosion can be defined as a metal removal process using a series of

electrical sparks to erode material from a work piece in a liquid medium under

carefully controlled conditions. It is a pressure-less process which does not trans-

mit heat to the alloy, making it possible to consistently machine fit in the range of

0.05 mm [10-51]. The procedure consists of a tension current passing through an

Fabrication Technologies 289

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 289

electrode, which produces a burning-away of minute particles. A graphite or cop-

per electrode acts as a tool that invades the piece of the base metal and erodes a

negative form in the shape of the electrode. The process is accompanied by sparks

generated between the electrode and the restoration. The sparks melt the alloy by

heating it to between 3000 and 5000

C [10-50, 10-52]. The sparks remove small

amounts of the metal substrate within microseconds. The entire process takes

place in a dielectric liquid bath that prevents the alloy from burning. Advantages

of this technique are (1) it is possible to work without pressure, (2) no heat is con-

ducted to the object being treated, so parts of the prosthesis which have already

been finished with porcelain or acrylic resin will not be damaged, (3) all parts can

be absolutely parallel, (4) very little surface roughness, and (5) no difficulties with

any type of material [10-50, 10-52, 10-53].

Anderson and Anderson [10-54] developed a method by which the principles

of spark erosion are combined with a method of machine duplication for CpTi

fabricating crowns. The Procera process developed by the Nobelpharm company

is based on a precision duplication milling through CAD/CAM, combined with

spark erosion and laser welding of the metal frame. Procera is designed for use

with titanium and densely sintered alumina making it possible to produce metal

ceramic crowns or all ceramic crowns [10-55]. An alternative to casting Ti for

ceramic applications was the development of CAD/CAM machining and spark

erosion of Ti [10-56, 10-57]. The main advantage of this technique is that it

overcomes the hardened surface layer that is encountered with Ti castings, there-

fore providing adequate porcelain–titanium bonding [10-58, 10-59]. It was men-

tioned that adaptation of the current Procera CAD/CAM method using

pressed-powder technology in the fabrication of Ti copings has several advan-

tages compared to the original Procera method [10-60, 10-61]. The original

method required at least four milling operations, the outer surface milling of the

Ti blank and milling of three electrodes for spark erosion of the inner form. The

powder method needs only two milling operations, one for the enlarged refrac-

tory die and one for the outer contour of the coping. Using the spark-erosion

process, the milling of the outer contour of the coping is performed on a Ti

blank. With the pressed-powder method, the powder is compacted onto the sur-

face of the refractory die and then milled before the sintering, which speeds up

the process and reduces wear of the milling tools. Although the spark-erosion

process requires careful alignment of the carbon electrodes with the Ti blank in

fabricating the coping, with powder technology, the alignment is not critical in

the same way. The spark-erosion process is a time-consuming and relatively

costly operation. The pressed-powder technology could become a very cost-

effective operation. The sintering process does take time; however, when the

process is used commercially in creating multiple copings at the same time, it

290 Bioscience and Bioengineering of Titanium Materials

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 290

could become cost effective. Several approaches have been reported in the liter-

ature using powder technology in dentistry [10-60, 10-61].

10.4. ISOTHERMAL FORGING

Isothermal forging differs from conventional forging in that the temperature of the

workpiece is held constant during forging by heating the dies. It is found that

the microstructure is affected by isothermal forging in various ways, the most impor-

tant being the production of a fine grained, equiaxed structure, and free of slip band-

ing. Therefore, by careful control of strain rate and temperature, control of the

microstructure, and hence the properties, can be achieved. Forging is a common

method of producing wrought titanium alloy articles. Forging sequences and subse-

quent heat treatment as post-forging heat treatment can be used to control the

microstructure and resulting properties of the product. Forging is more than just a

shape-making process. The key to successful forging and heat treatment is the beta

() transus temperature (i.e., 885°C). The higher the processing temperature in the

( + ) region, the more beta is available to transform on cooling. The solution heat

treatment offers a chance to modify or tune the as-forged microstructure, while the

age cycle modifies the transformed beta structures to an optimum dispersion [10-62].

Titanium alloys are known to be very susceptible to their metallurgical back-

ground, and their properties can vary considerably depending on the alpha (HCP)

↔ beta (BCC) transformation temperature. Microstructural control is basic to the

successful processing of titanium alloys. Undesirable structures (grain-boundary

alphas, beta fleck, elongated alpha) can interfere with optimum property develop-

ment. Titanium ingot structures can carry over to affect the forged product. Beta

processing, despite its adverse effects on some mechanical properties, can reduce

forging costs, while isothermal forging offers a means of reducing forging pres-

sure and/or improving die fill and part detail. The first step is to heat the tooling

to the forging temperature. A good deal of progress has been made in this area and

tooling sets are now available that can withstand elevated temperatures. A second

factor for the success of isothermal forging is the forgeability of the alloy: the

lower the temperature needed to forge the alloy, the easier it will be to form at uni-

form temperature. This is why isothermal forming is easier with  or near  alloys,

which have a much lower -transus and lower resistance to hot forming [10-62].

10.5. SUPERPLASTIC FORMING (SPF)

The phenomenon of superplasticity was first observed over 70 years ago, but

active research in this field did not begin until 1960. Since then, superplastic

Fabrication Technologies 291

Else_BBTM-OSHIDA_CH010.qxd 9/14/2006 6:32 PM Page 291