WO2023021418A1

WO2023021418A1 - Very high strength copper-titanium alloy with improved formability in the solution annealed temper

Info

Publication number: WO2023021418A1
Application number: PCT/IB2022/057654
Authority: WO
Inventors: Baptiste ROUXEL
Original assignee: Ln Industries Sa
Priority date: 2021-08-17
Filing date: 2022-08-16
Publication date: 2023-02-23
Also published as: CN117795109A

Abstract

The copper-titanium alloy according to the invention comprises, by weight, at least 90% copper, from 5 to 7% titanium and from 0.25 to 0.5% iron. It has excellent ductility at the solution-annealed temper and high yield strength after an ageing treatment.

Description

Very high strength copper-titanium alloy with improved formability in the solution annealed temper

The present invention relates to a copper alloy that can be used as a substitute for copper-beryllium (Cu-Be) alloys.

The Cu-Be alloys are widely used in many applications where they are particularly appreciated for the combination of their high electrical conductivity and very good mechanical properties. They combine a temper (solution-annealed (SA) temper) with a very high ductility allowing extreme formability (elongation superior to 40%) and another temper (age-hardened (AH) temper) with an extremely high mechanical resistance (over 1 GPa). Nevertheless, they are doomed to disappear as Be is highly toxic and the alloys using this element are progressively forbidden.

Copper-titanium (Cu-Ti) alloys represent good alternative alloys to Cu-Be. Like Cu-Be alloys, the Cu-Ti alloys are age-hardenable alloys, i.e. , it is possible, under some specific conditions, to dissolve the alloying element Ti in the solid copper phase which has a face-centered cubic lattice. At this temper, often referred to as the solution-annealed temper (SA), the material provides usually the maximum formability. When the Cu-Ti alloys are aged at an intermediate temperature, usually between 300°C and 550°C, the Ti precipitates into a finely and uniformly dispersed a-Cu4Ti tetragonal nano-metastable phase (from 20 to 200 nm large), which considerably increases the strength of the alloy. If the alloy is maintained longer in temperature, larger orthogonal stable phases, referred to as [3-Cu4Ti or CusTi in the literature, form as cells on the grain boundaries and are detrimental for the mechanical properties of the alloy. This phenomenon is called “over-ageing”.

Most of current developments aim at hindering the formation of CusTi and/or increasing the electrical conductivity of Cu-Ti alloys for electric applications. To solve any of these challenges, the content of Ti is often kept under 4 wt.%.

US patent application No. 2004/0136861 discloses a copper alloy for use in connector materials and intended to have excellent bendability and to be preserved from the precipitation of CusTi. This copper alloy contains from 2 to 4 wt.% of Ti and from 0.01 to 0.5 wt.% of at least one element selected from Fe, Co, Ni, Cr, V, Zr, B and P as a third element group, wherein not less than 50% of the total content of the third element group exists as a second-phase particle.

The present invention focuses on all the applications of Cu-Be that do not require good electrical conductivity but only high mechanical resistance. This allows adding a larger amount of Ti, usually detrimental to electrical conductivity. It is known that when Ti is added in the alloy in high amount (over 4% by weight), the Cu-Ti alloys exhibit a yield strength comparable with Cu-Be but suffer from a low ductility in the solution-annealed temper due to the difficulty of keeping the Ti in solid solution during the water quench which ends the solution annealing. For example, in a study on the effect of increasing the amount of Ti, S. Nagarjuna et al. obtained an elongation of only 23% with a Cu-Ti alloy containing 5.4 wt.% of Ti (see the article “On the variation of mechanical properties with solute content in Cu-Ti alloys”, by S. Nagarjuna et al., Materials Science and Engineering A259 (1999) 34-42).

US patent No. 4,599,119 discloses an age-hardened copper-titanium alloy containing from 2 to 6%, preferably from 3 to 5%, by weight of titanium and having an average crystal grain size not greater than 25 microns, preferably between 3 and 15 microns. Besides copper and titanium, this alloy may contain at least one element among iron, zirconium, chromium, boron and silicon, not exceeding 2% by weight in a total amount. In this document, the small crystal grain size is considered to improve mechanical properties of Cu-Ti such as isotropy, formability, fatigue strength, elongation and yield strength. The small crystal grain size is achieved through appropriate heat treatments, including a pre-annealing at intermediate temperatures to form spherical precipitates mentioned as a secondary phase. No particular effect is mentioned with regard to the possible use of iron, zirconium, chromium, boron or silicon. This patent does not address the problem caused by high amounts of titanium (over 4%) to the ductility of the alloy. US patent No. 2,783,143 discloses an age-hardenable copper-base alloy having improved strength and ductility in the age-hardened temper. This alloy contains from 1 to 10% titanium, from 0.1 to 1.6% cobalt, from 0.05 to 0.8% chromium, from 0.04 to 0.62% nickel, from 0.04 to 0.60% iron, from 0.02 to 0.28% molybdenum and from 0.005 to 0.08% manganese. Preferably, this alloy contains from 2 to 6% titanium, from 0.2 to 0.8% cobalt, from 0.1 to 0.4% chromium, from 0.08 to 0.31 % nickel, from 0.075 to 0.3% iron, from 0.035 to 0.14% molybdenum and from 0.01 to 0.04% manganese. Although not mentioned, the indicated percentages seem to be atomic percentages. In the examples shown, the content of titanium is 4%, which corresponds to about 3 wt.%. This patent focuses on the ductility in the age-hardened temper and does not mention the ductility before the age hardening, i.e., in the solution-annealed temper aiming at providing the maximum formability. Moreover, this patent does not address the problem caused by high amounts of titanium (over 4 wt.%) to the ductility of the alloy.

International patent application No. WO 2021/143257 discloses a titanium bronze alloy for explosion-proof tools. The alloy comprises, by weight, from 5 to 7% titanium, from 0.8 to 1.5% aluminum, from 0.1 to 0.3% silver, from 0.2 to 0.4% iron, from 0.03 to 0.08% rare earth, the balance being copper. This document does not mention the ductility in the solution-annealed temper. Moreover, the alloy proposed is expensive due to the presence of silver.

Japanese patent application No. JP 2021/050393 discloses a titanium copper alloy plate for a vapor chamber. According to one example, the alloy comprises, by weight, 4.8% of titanium and 0.2% of iron. A comparative example is mentioned which comprises, by weight, 5.2% of titanium and 0.2% of iron. The alloy in this document does not have an optimum ductility in the solution-annealed temper.

An object of the present invention is to provide a copper-titanium alloy which can have both an excellent ductility in the solution-annealed temper and a high yield strength after an ageing treatment. To this end, there is provided a copper-titanium alloy comprising, by weight, at least 90% copper, from 5 to 7% titanium and from 0.25 to 0.5% iron.

The content of titanium, by weight, is preferably at least 5.2%, preferably at least 5.5%, preferably at least 6%, and is preferably at most 6.5%.

The content of iron, by weight, is preferably at most 0.4% and preferably at most 0.35%.

The copper-titanium alloy may further comprise aluminum at a content of not more than 1.4% by weight, preferably at a content of from 0.1 to 1.4% by weight, preferably at a content of from 0.1 to 0.7% by weight, preferably at a content of from 0.1 to 0.6% by weight.

Preferably, the copper-titanium alloy comprises no or little silver, i.e. , the content of silver in the alloy is at most 0.08%, preferably at most 0.07%, preferably at most 0.06% by weight.

The present invention further provides a method for manufacturing the copper-titanium alloy as defined above, with improved formability, comprising a solution annealing step comprising a heat treatment performed at a temperature of at least 840°C followed by a quench (fast cooling).

Other features and advantages of the present invention will be apparent upon reading the following detailed description made with reference to the appended drawings in which:

- Figure 1 is a graph showing the elongation before ageing (at the solution- annealed (SA) temper) as a function of the yield stress (0.2% offset) after ageing (age-hardened temper). The data of Cu-XTi-0.3Fe alloys (with the nominal value X being equal to 3, 5 or 6) are superimposed with the commercial data of industrial CuBe-C17200 alloy and C72900 alloy at different tempers;

- Figure 2 is a graph showing the evolution of the hardness in Vickers of the alloys Cu-6Ti and Cu-6Ti-0.3Fe over an ageing at 450°C from the solution-annealed temper; - Figure 3 is a graph showing the stress-strain engineering tensile curves of the alloys Cu-6Ti and Cu-6Ti-0.3Fe at the solution-annealed (SA) temper and at a predetermined peak-aged condition (2h at 450°C) aiming at providing the maximum strength maintaining reasonable ductility in the age-hardened (AH) temper;

- Figure 4 is a graph showing the evolution of the yield stress with the Ti content in the alloy, at the solution-annealed (SA) temper and at the age- hardened (AH) temper (2h at 450°C);

- Figure 5 is a graph showing the evolution of the elongation at failure with the Ti content in the alloy, at the solution-annealed (SA) temper and at the age-hardened (AH) temper (2h at 450°C);

- Figure 6(a) is a scanning electron microscopy (SEM) micrograph of Cu- 6Ti at the solution-annealed (SA) temper after chemical electro-polishing where the precipitates have consequently been exposed and are visible with a secondary electron detector. The hardness was measured at 320 Hv;

- Figure 6(b) is a scanning electron microscopy (SEM) micrograph of Cu- 6Ti-0.3Fe at the solution-annealed (SA) temper after chemical electropolishing where the precipitates have consequently been exposed and are visible with a secondary electron detector. The hardness was measured at 160 Hv;

- Figure 7 is a transmission electron microscopy (TEM) taken along the zone axis B=(100) of the alloys (a) Cu-6Ti and (b) Cu-6Ti-0.3Fe at the solution-annealed (SA) temper, with (c) and (d) being the corresponding diffraction patterns;

- Figure 8 shows an X-ray diffraction (XRD) of Cu-6Ti (top curve, T6) and Cu-6Ti-0.3Fe (bottom curve, T7) in the solution-annealed temper. The top right windows I, II, III are three relevant zooms in the spectrum; - Figure 9 shows an X-ray diffraction (XRD) of three different alloys (Cu- 6Ti-0.1 Fe; Cu-6Ti-0.3Fe; Cu-6Ti-0.7Fe) in the solution-annealed temper, zoomed in the angle 20 corresponding to the diffraction of the plane <100> of Cu;

- Figure 10 is a scanning electron micrograph (SEM) of the alloys Cu-6Ti- 0.3Fe and Cu-6Ti-0.7Fe at the solution-annealed temper;

- Figure 11 shows the hardness of three alloys (Cu-6Ti-0.3Fe; Cu-6Ti- 0.3Fe-0.2Ag; Cu-6Ti-0.3Fe-1 .2AI) at the solution-annealed temper and at the age-hardened temper.

Copper-titanium alloys have been studied in the last century. They are conventionally manufactured by implementing the following sequence: a) Casting: the elements of the alloy are mixed at a high temperature in the liquid phase. An induction furnace is preferably used. The temperature is held for at least 1 min over the liquidus temperature (which is around 1000°C), preferably between 1200°C and 1300°C. The alloy is preferably cast in a mold to guarantee a faster cooling and better homogeneity of the melt. Depending on the Cu-Ti alloy, strategies can be used to introduce the elements in a specific order and/or from master alloys. b) Homogenization: a homogenization heat treatment is necessary to homogenize the Ti that segregates during the solidification process along the dendrites. It must be carried out in between 880°C and 950°C (under the solidus temperature) and can last in between 1 h and 48h depending on the size of the cast. The material is then cooled relatively slowly, preferably in a neutral atmosphere to avoid too much oxidation. c) Forming (preferred but optional): a hot deformation in between 750°C and 900°C and/or a cold deformation can be performed to reach the final shape of the piece for a given application. Hot deformations increase the homogeneity of the microstructure with smaller grains and allow large deformations, while cold deformations give a better control of the geometry. d) Solution annealing (SA): the alloy is heated and kept at a high temperature for a certain time in order to dissolve the titanium in the solid phase to form a solid solution. Then the alloy is quickly cooled by water quench to freeze the microstructure. The temper of the alloy after the water quench is called “solution-annealed temper” or “SA temper”. e) Ageing or “age hardening”: the ageing is usually performed at a temperature comprised between 300°C and 550°C depending on the precold work, to precipitate the titanium and provide the maximum yield stress. The temper of the alloy after the ageing is called “age-hardened temper” or “AH temper”.

At the end of the solution annealing, the alloy is soft and can thus be easily deformed to take its final shape. At the end of the ageing step, the alloy has a higher yield strength, but the latter depends on the content of Ti. It has been shown that increasing the content of Ti up to 7 wt.% increases the strength of the alloy in the age-hardened temper. However, it was then observed that with more than 4 wt.% Ti in the alloy, it is impossible to keep the Ti in solid solution after the water quench. This results in a wave-like microstructure observed in TEM (transmission electron microscopy). The mechanism behind this observation is controversial but always relates to an early-stage precipitation, resulting in a large increase in strength and a decrease in the ductility. This behavior is problematic, as it makes it impossible to produce a binary Cu-Ti alloy with high strength at the age-hardened temper (more than 850 MPa) and with high formability (elongation superior to 30%) at the solution- annealed temper.

The present invention is based on the observation that small additions of Fe (0.25 to 0.5 wt.%, preferably 0.25 to 0.4 wt.%, preferably 0.25 to 0.35 wt.%) combined with a high content of Ti (5 to 7 wt.%, preferably 5.2 to 7 wt.%, preferably 5.5 to 7 wt.%, preferably 5.5 to 6.5 wt.%) results in suppressing this early-stage precipitation and keeping the titanium in solid solution. This allows doubling the elongation of the alloy in the solution-annealed temper, reaching then more than 40% and even more than 50% plastic deformation (elongation), while keeping, or even increasing, the yield strength of the alloy in the age-hardened temper (more than 900-1000 MPa). For reaching higher and more isotropic mechanical properties, the grain size is preferred to be inferior to 30 pm.

The content of Cu in the alloy of the invention is at least 90 wt.%. Excellent results are obtained when only Cu, Ti and Fe (and unavoidable impurities) are present in the alloy. Addition of one or more other elements such as Co, Zr, Si, P, Ni, Sn, Zn, Pb, Mn, Mg, As, Sb and Cr is not excluded in the invention but no improvement in terms of ductility in the solution-annealed temper and yield strength in the age-hardened temper has been observed with such added elements. A deterioration of the results with respect to Cu-Ti-Fe may even occur if the added element, e.g. Co, reacts with Fe. The technical effect in the invention indeed relies on the action of Fe combined with a high content (5-7 wt.%) of Ti. Specifically in this high range of Ti, small additions of Fe increase considerably the ductility of the alloy. Unlike the aforementioned other elements, Fe has a special effect on Ti precipitation and ductility. Its role is expected to be in solid solution or through the formation of new nano-precipitates but not through the formation of secondary phases as described in US 2004/0136861.

Given the high content of titanium in the invention, the solution heat treatment of the solution annealing (step d) above) is performed at a temperature of at least 840°C, preferably of at least 850°C, preferably of at least 880°C.

Detailed test results will now be set out with reference to the appended figures. The Cu-Ti-Fe alloys on which these results are based were manufactured as follows. Different model alloys T1 , T2, T3, T4, T6 and T7 were casted from high purity copper metal, iron and titanium sponges. About 500g of metal were weighted and melted in a sealed induction furnace under argon gas. After holding the temperature during 5 min at 1250 °C the alloys were poured in a graphite crucible to form 150 x 50 x 11 mm ingots. The resulting chemical compositions were analyzed in the middle of each ingot by Inductively Coupled Plasma - Atomic Emission Spectrometry (ICP-AES) and are shown in the table below, in weight percentages:

The indications “<0.01 ” mean that the presence of the corresponding element in the alloy is not measurable. The Si in the alloys T1 to T7 is an impurity. Likewise, the P in the alloy T2 is an impurity. The thermomechanical treatments were adapted to the amount of Ti for each alloy. The ingots were homogenized to get rid of the solidification chemical segregations, and then were hot rolled (HR) at 850°C under air, reducing the plates from 11 mm to 3.6 mm thick in three successive passes, in order to ensure a good dynamic recrystallization and small equiaxed grains. The alloys were successively cold rolled (CR), solution annealed with a flux of argon and water quenched (SA). These processes were repeated a large number of times to reach a thickness of 0.5 mm. The SA temperature and CR were tailored to reach a grain size under 60 pm. A grain size under 30 pm is preferred to reach better and more isotropic mechanical properties. These samples are referred to as the SA tempers in the following. Some of the specimens were then aged at 450°C, under Ar gas during 2h to produce the temper referred to as AH (age-hardened) temper.

Figure 1 shows a graph aiming at demonstrating the main benefits of the Cu- 6Ti-0.3Fe alloy (alloy T7) in comparison with the Cu-Be (C17200), the binary Cu-Ti (alloys without iron T1 , T3 and T6) and a standard alternative alloy to Cu-Be (C72900). Thanks to the micro-addition of Fe, Cu-Ti(-Fe) alloys combine good formability before ageing and very high yield strength after ageing.

More particularly, Figure 1 shows the elongation before ageing (after quench; solution-annealed temper) on the y-axis, a value related to the formability of the alloys at this temper, and the x-axis gives the yield stress of the alloys after ageing, a value related to its resistance in working condition. This graph is relevant for any application, such as the Bourdon tubes in manometers, which requires a very high formability to manufacture products with complex shapes and high mechanical properties in the working conditions. The Cu-Be alloy (represented by squares in the graph) is by far the best in this dual behavior. However, the addition of 0.3 wt.% of Fe to Cu-Ti alloys (empty circles), with a content of Ti of 5 or 6 wt.%, allows to approach the behavior of Cu-Be and this, much more than the commercial C72900 alloy indicated in the graph by the black triangles.

Figure 2 shows the evolution of the hardness of the Cu-6Ti (alloy T6) and the Cu-6Ti-0.3Fe (alloy T7) as a function of the ageing time at 450°C. It can be observed that the addition of 0.3 wt.% of Fe into a Cu-6Ti alloy first increases the maximum hardness from 330 Hv to 350 Hv but especially stabilizes this hardness at 450°C. This shows that the micro-addition of Fe limits the over-ageing, which is one of the main challenges in Cu-Ti alloys, always associated in the literature with the formation of the stable phase [3-Cu4Ti (or CusTi) at the grain boundaries. Figure 3 shows that Cu-6Ti-0.3Fe (alloy T7) is softer than Cu-6Ti (alloy T6) at the solution-annealed (SA) temper and harder than Cu-6Ti at the age-hardened (AH) temper. This property can be very appreciated in the watch industry, for example. In the solution-annealed temper, with the addition of only 0.3 wt.% of Fe in a Cu-6Ti, the elongation is multiplied by a factor 2 and the yield stress is almost divided by 2. Such a tremendous effect was not expected. It can be explained by the absence of spinodal decomposition in the solution-annealed temper, made possible by the micro-additions of Fe.

The Cu-6Ti-0.3Fe also exhibits a very profitable balance between yield strength and elongation at the age-hardened temper. This balance between strength and formability is superior to the one in the commercial Cu-Be alloys and the standard commercial Cu-15Ni-8Sn alternative. Unlike comparable commercial alloys on the market, the Cu-6Ti-0.3Fe in the age-hardened state can reach a yield stress superior to 1 GPa with an elongation superior to 15%.

Figures 4 and 5 respectively show the evolution of the yield stress and of the elongation at failure as a function of the Ti content in the alloy, at the solution- annealed temper and the age-hardened temper, for the alloys Cu-Ti and Cu-Ti-Fe (0.3 wt.% Fe) processed as described above and for the binary Cu-Ti alloys with a grain size of 60 pm disclosed in the article “On the variation of mechanical properties with solute content in Cu-Ti alloys” by S. Nagarjuna et al., Materials Science and Engineering A259 (1999) 34-42, and processed as described in the said article. These figures illustrate the need of having a high Ti content (5-7 wt.%) to observe significantly the described benefits of Fe micro-additions.

The physical mechanism in which micro-additions of Fe are effective is illustrated in Figures 6 and 7. The nano-precipitation in the alloy Cu-6Ti (alloy T6) is visible in Figure 6(a) and, by the very specific wave-like aspect along the direction <200>, in Figure 7(a). It can be seen from Figures 6(b) and 7(b) that the microadditions of Fe in the alloy completely suppress such early precipitation. Figure 8 shows an X-ray diffraction (XRD) of Cu-6Ti (top curve, alloy T6) and Cu-6Ti-0.3Fe (bottom curve, alloy T7) in the solution-annealed temper. It confirms the assumptions of the mechanism explained above. The early-stage precipitation associated sometimes with spinodal decomposition is known to create a sideband effect in XRD. This effect is clearly visible in Figure 8 on the curve obtained from the alloy T6. On the curve obtained from the alloy T7, the peaks corresponding to Cu are much thinner, because of the absence of spinodal decomposition. This observation is in good agreement with Figures 6 and 7. The zoom I shows the typical broadening of the peak of Cu due to spinodal decomposition in the binary Cu-6Ti. This broadening is absent in the Cu-6Ti-0.3Fe, revealing the effect of Fe in stopping the precipitation during the water quench.

To avoid spinodal decomposition, it is required that the content of Fe be sufficiently high. Figure 9 shows X-ray diffraction (XRD) results for three different alloys (Cu-6Ti-0.1 Fe; Cu-6Ti-0.3Fe (alloy T7); Cu-6Ti-0.7Fe) at the solution- annealed temper. The shoulders visible on both sides of the diffraction peak and known in the literature as shoulder effect are a proof of the presence of spinodal decomposition. One finding of the present invention is to add Fe to suppress this spinodal decomposition at the solution-annealed temper and keep the alloy ductile. The results show that the spinodal decomposition appears when the Fe content is too low (0.1 wt.% Fe) but disappears when the Fe content is higher (0.3 wt.% Fe). 0.3 wt.% Fe seems to be the optimum concentration to efficiently get rid of this spinodal decomposition and associated shoulder effect on the diffraction peak.

The content of Fe must however not be too high since otherwise large undesired TiFe intermetallics are formed which have a negative impact on the ductility and mechanical properties of the alloy. Such large intermetallics are visible in Figure 10 for Cu-6Ti-0.7Fe.

As already mentioned, the addition of one or more elements in the alloy according to the invention is not excluded. Figure 11 shows the hardness of the Cu- 6Ti-0.3Fe in the solution-annealed temper and in the age-hardened temper without Ag nor Al in the alloy (left-hand graph), with 0.2 wt.% Ag in the alloy (middle graph) and with 1.2 wt.% Al in the alloy (right-hand graph). Unexpectedly, Al softens the alloy in the solution-annealed temper, thus improving the ductility in that temper, without decreasing the hardness in the age-hardened temper. The content of Al must however be sufficiently low to avoid the formation of intermetallics with Fe. Preferably, aluminum is present in the alloy at a content of from 0.1 to 1.4 wt.%, preferably of from 0.1 to 0.7 wt.%, preferably of from 0.1 to 0.6 wt.%. As to Ag, the graph of Figure 11 shows that it has no positive effect on the mechanical properties of the alloy (the hardness in the solution-annealed temper is not decreased and the hardness in the age-hardened temper is not increased). In view of the very high cost of Ag, the alloy according to the invention preferably includes no or little Ag, i.e. , the content of silver in the alloy is at most 0.08 wt.%, preferably at most 0.07 wt.%, preferably at most 0.06 wt.%. An alloy comprising only Cu, Ti, Fe and Al (and unavoidable impurities) can therefore be particularly advantageous.

The alloy according to the invention can replace the Cu-Be alloys in any very high strength alloy application requiring good mechanical properties and where a good electrical conductivity is not needed, for example the Bourdon tubes in high- pressure manometers, parts of timepieces (e.g., watch cases, gears, escapements, balances, springs, shafts, oscillating weights, plates, bridges, hands, dials, discs, etc.), ball-roller bearings and bushings, in particular for the aircraft and airspace industry, and dies for plastic extrusion.

Claims

CLAIMS Copper-titanium alloy comprising, by weight, at least 90% copper, from 5 to 7% titanium and from 0.25 to 0.5% iron. Copper-titanium alloy according to claim 1 , wherein the content of titanium, by weight, is at least 5.2%. Copper-titanium alloy according to claim 1 , wherein the content of titanium, by weight, is at least 5.5%. Copper-titanium alloy according to claim 1 , wherein the content of titanium, by weight, is at least 6%. Copper-titanium alloy according to any of claims 1 to 4, wherein the content of titanium, by weight, is at most 6.5%. Copper-titanium alloy according to any of claims 1 to 5, wherein the content of iron, by weight, is at most 0.4%. Copper-titanium alloy according to any of claims 1 to 5, wherein the content of iron, by weight, is at most 0.35%. Copper-titanium alloy according to any of claims 1 to 7, including only copper, titanium, iron and unavoidable impurities. Copper-titanium alloy according to any of claims 1 to 7, further comprising aluminum at a content of not more than 1 .4% by weight. Copper-titanium alloy according to any of claims 1 to 7, further comprising, by weight, from 0.1 to 1.4% aluminum. Copper-titanium alloy according to any of claims 1 to 7, further comprising aluminum at a content of from 0.1 to 0.7%, preferably of from 0.1 to 0.6% by weight. Copper-titanium alloy according to any of claims 9 to 11 , including only copper, titanium, iron, aluminum and unavoidable impurities. Copper-titanium alloy according to any of claims 1 to 7 and 9 to 11 , comprising no silver or comprising silver at a content of not more than 0.08%, preferably 0.07%, preferably 0.06% by weight. Use of the copper-titanium alloy according to any of claims 1 to 13 for manufacturing a Bourdon tube. Use of the copper-titanium alloy according to any of claims 1 to 13 for manufacturing a part of a timepiece. Method for manufacturing a copper-titanium alloy according to any of claims 1 to 13, comprising a solution annealing step comprising a heat treatment performed at a temperature of at least 840°C, followed by a quench.