Titanium and its alloys are chosen for airframes, engine components, chemical plant and orthopaedic implants because few other metals offer the same balance of strength, density and corrosion resistance (Lütjering and Williams, 2007). The same alloys have a poor reputation wherever surfaces must slide. Under unlubricated contact titanium tends to adhere to the opposing body, to transfer material and to generate loose debris, and designers have usually responded by coating, nitriding or otherwise modifying the surface instead of relying on the bulk alloy (Budinski, 1991).

Additive manufacturing (AM) has become an established route for titanium components over the past two decades. Powder bed fusion with a laser or an electron beam, and directed energy deposition, build parts layer by layer from powder or wire, which suits an expensive metal that is difficult to machine (Herzog et al., 2016). The thermal history of these processes differs sharply from that of casting or forging. Melt pools are small, cooling is rapid and each layer is reheated by those deposited above it, so the resulting microstructures, residual stresses and defects have no close equivalent in conventionally processed stock (DebRoy et al., 2018).

Mechanical properties of AM titanium have been reviewed many times, with attention concentrated on tensile and fatigue behaviour (Lewandowski and Seifi, 2016; Liu and Shin, 2019). Tribological behaviour has received less systematic treatment, although AM parts increasingly include bearing surfaces, threads, sliding joints and implant interfaces. This article reviews what is known about friction and wear of AM titanium alloys in dry sliding, and asks whether the change of manufacturing route alters the long-standing picture of titanium as a metal that wears badly.

The article is a narrative review of the published literature. It draws on established texts and reviews of titanium metallurgy and of AM, on the classical studies of titanium wear, and on experimental papers that compare AM titanium with cast or wrought material in unlubricated sliding. No formal search protocol was applied and no claim of completeness is made. Coverage is limited to commercially pure titanium (CP-Ti) and Ti-6Al-4V, which account for most of the reported work, and to sliding in ambient air at room temperature. Lubricated sliding, fretting, tribocorrosion in simulated body fluid and titanium matrix composites are mentioned only where they clarify the dry sliding case.

Any assessment of AM titanium needs the behaviour of wrought material as its reference. The classical account attributes the poor wear resistance of titanium to a combination of properties. The metal has a low resistance to plastic shearing and work hardens little, so surface layers deform readily under tangential load. Its native oxide film is thin and is easily broken, exposing reactive metal that bonds to the counterface. Its thermal conductivity is low, so frictional heat is concentrated at the interface and the contact temperature rises faster than it would on steel or aluminium under the same conditions (Budinski, 1991; Hutchings and Shipway, 2017).

Molinari et al. (1997) examined Ti-6Al-4V in dry sliding and described two regimes: oxidative wear at low sliding speeds, where a brittle surface oxide forms and spalls without protecting the metal, and delamination at higher speeds, where subsurface plastic deformation leads to the detachment of metallic flakes. Straffelini and Molinari (1999) showed that the outcome also depends on the counterface and on the combination of load and speed. Work with metallic, polymeric and ceramic counterfaces has confirmed that transfer of titanium to the opposing surface is common and that friction traces are often unsteady (Qu et al., 2005). Against hardened steel, increasing the sliding speed can change both the wear rate and the dominant mechanism (Alam and Haseeb, 2002).

These observations sit within general frameworks of sliding wear. The Archard relation takes wear volume to be proportional to load and sliding distance and inversely proportional to hardness (Archard, 1953), and wear-mechanism maps set out how the governing mechanism shifts with contact pressure and velocity (Lim and Ashby, 1987). On many alloys a compacted, oxidised layer of debris forms at elevated interface temperature and lowers wear (Stott, 1998). On titanium the mechanically mixed layer produced by transfer and back-transfer (Rigney, 2000) is generally reported to be poorly adherent at room temperature, which is one reason why hardness alone predicts titanium wear less well than the Archard relation would suggest.

The features of AM titanium that matter for sliding contact can be grouped under microstructure, defects and surface condition. Figure 1 summarises how they differ between the main process routes and the conventional reference.

Experimental comparisons of AM and conventional titanium share a common design: flat specimens cut from AM builds and from cast or wrought stock are polished and slid against a harder ball or pin, and wear volume, friction coefficient and worn surface morphology are compared. Table 1 gathers the main factors that these studies and the wider literature identify, with the direction of their reported or expected effect.

Most of the studies cited use a pin-on-disc or ball-on-flat arrangement of the kind standardised in ASTM G99 (ASTM International, 2017) or a linearly reciprocating ball-on-flat test as in ASTM G133 (ASTM International, 2016). Within one study the conditions are held fixed and the ranking of materials is meaningful. Between studies almost everything varies: the counterface may be bearing steel, alumina, silicon nitride or cemented carbide; loads, speeds and sliding distances differ by large factors; humidity is often unreported; and wear is expressed variously as mass loss, volume loss, track width or specific wear rate.

Since titanium wear is sensitive to the position of the test on the load and speed map (Lim and Ashby, 1987; Straffelini and Molinari, 1999), absolute wear rates from different laboratories cannot be pooled, and statements such as "AM titanium wears less than wrought titanium" are safe only as summaries of within-study rankings. The description of the AM material is often incomplete as well. Density, heat treatment, build orientation of the tested face and the depth of material removed in preparation all bear on the result and are not always given.

Several questions follow from the state of the literature. The first concerns the as-built surface. Parts with internal channels or lattice regions cannot be machined everywhere, and the sliding behaviour of a surface carrying sintered particles and a different near-surface microstructure is largely uncharacterised. The second concerns heat treatment. Service parts are rarely used as built, yet the comparative wear studies mostly test the as-built condition, in which hardness is highest. Whether the advantage over cast material survives stress relief or hot isostatic pressing has seldom been tested directly.

A third question is anisotropy. Columnar prior-beta grains and layered defects make it plausible that wear depends on whether sliding takes place on a plane parallel or perpendicular to the build direction, but the published findings do not agree. A fourth is the interaction between AM substrates and the surface treatments that titanium needs in any serious tribological duty. Thermal oxidation, for example, is known to improve the wear resistance of conventional titanium markedly (Dong and Bell, 2000), and the response of a martensitic, residually stressed substrate to such treatments deserves systematic study. Finally, the field would benefit from a small number of agreed test conditions and reporting items so that results from different laboratories could be set side by side.

The published evidence supports a measured view. Additively manufactured CP-Ti and Ti-6Al-4V wear in dry sliding by the same combination of adhesion, abrasion, delamination and oxidation as their cast and wrought counterparts. Where direct comparisons exist, the finer and harder structures produced by powder bed fusion usually give wear resistance that matches or modestly exceeds that of the conventional alloy, provided the material is dense. Friction is essentially unchanged. The underlying weaknesses of titanium as a sliding material, a non-protective oxide and a strong tendency to transfer, are properties of the metal and not of the manufacturing route, so surface engineering remains necessary. Firmer conclusions await studies that test heat-treated and as-built surfaces, report the material condition fully and use comparable test conditions.

To be declared by the author(s) before publication.

To be declared by the author(s) before publication.