GB2575365A - Process - Google Patents

Abstract

A chemical etching method for improving the surface smoothness of 3D printed titanium and titanium alloy articles is described. The titanium surface is produced via additive manufacturing. The method of improving the surface of the articles is achieved through immersion of the articles in an etching solution where the etching solution is maintained at a temperature of 2-85 oC. The etching solution comprises 20-45 v/v% nitric acid, 4-14 wt% ammonium bifluoride (ammonium hydrogen fluoride). Electron beam (EB) manufacturing or selective laser sintering (SLS) is used to produce the article and it may be etched until between 125-750µm of surface material has been removed.

Description

Process

Field of Invention

The invention relates to a chemical etching method for improving the surface smoothness of 3D printed titanium and titanium alloy articles. Excellent surface smoothness is obtained without compromising on other factors, such as hydrogen embrittlement, fatigue properties, excessive material removal, safety or environmental factors.

Background to the Invention

Additive manufacturing (also known as 3D printing) is becoming an important fabrication technique and can be used for manufacturing an increasing number of materials. One important use is the additive manufacture of articles made from titanium and titanium alloys, which is important for the manufacture of parts for many industries including the aerospace and healthcare industries. Additive manufactured (also known as 3D printed) parts can be highly customised without needing to prepare customised manufacturing machinery for each different job. The parts can also have complex 3D geometries that are not normally accessible with conventional manufacturing machinery.

Additive manufacture of titanium and titanium alloys is typically done using powderbed fusion. A laser or electron beam is used to melt and fuse the required areas of successive layers of metal powder together to construct the required 3D shape. The unfused powder is then removed to leave the fused article.

One drawback of additive manufacturing is that the surfaces of the finished article are rough. For example, laser fusion techniques typically produce surfaces with an Ra value of about 10 to 20 pm and electron beam techniques typically produce even rougher surfaces, having Ra values of about 20 to 30 pm.

Surface roughness of this magnitude is not acceptable in many applications. Surface roughness can provide unwanted friction and is particularly problematic on moving parts or aerodynamic surfaces requiring low drag. Furthermore, the rough edges are typically weak and can provide stress concentration points, leading to poor fatigue performance.

Chemical etching has been applied to titanium and titanium alloy materials produced by conventional manufacturing techniques, for example for removal of surface oxides. Chemical etching has also been employed in surface treatment of 3D printed titanium and titanium alloy articles. There are, however, many factors that need to be taken into account when using chemical etching. This includes the hazardous nature of the etchants, which typically comprise toxic and/or corrosive acids such as hydrochloric acid, sulfuric acid, hydrofluoric acid and nitric acid. Chemical etching also produces significant quantities of hydrogen gas at the surface of the article, which results in the formation of titanium hydrides and trapping of gas bubbles at the surface. The formation of titanium hydrides leads to unwanted embrittlement at the surface of articles. The trapping of gas bubbles at the surface leads to pitting and voids. These both adversely affect the fatigue behaviour of components.

A need remains to improve the surface properties of 3D printed titanium and titanium alloy articles when treated using chemical etching.

Summary of the Invention

According to a first aspect, the invention provides a method of increasing the surface smoothness of an additive manufactured titanium or titanium alloy article, the method comprising immersing the article in an etching solution, wherein the etching solution flows relative to the article, and wherein the etching solution comprises 20-45 v/v% nitric acid and 4-14 wt% ammonium bifluoride, and the etching solution is maintained at a temperature between 2 and 85 °C.

More specifically, the nitric acid is a commercially available nitric acid, which is a solution of 68 wt% nitric acid in water, and the etching solution is aqueous.

There are many advantages associated with this method, as set out below.

These chemical etching conditions allow for the surface smoothness of 3D printed titanium and titanium alloy articles to be increased to exceptional levels. Unlike with mechanical milling, this surface smoothness can be generated on all surfaces that can be reached by the etchant solution, and not just the surfaces that can be reached by mechanical milling tools. As a result of this, surface friction, fatigue performance and appearance are all greatly improved across the entire article. The method can be applied with equal effect to articles produced by either electron beam or laser additive manufacture.

Furthermore, it has been shown that the surfaces produced by this method are smooth enough to be examined by non-destructive testing techniques such as liquid penetrant inspection (LPI). LPI is described in ASTM E1417-99, and is essentially based on rough surfaces trapping and retaining a penetrant dye more than smooth surfaces. LPI is used for the detection of cracks and other defects in the surface of components, but surface roughness generates background noise that compromises the ability to detect other defects. Smoother surfaces allow for more sensitive liquid penetrants to be utilised, which results in a greater likelihood of small cracks and defects being detected. In addition to the surfaces produced by the present invention being smooth enough to use LPI, the surfaces are even smooth enough that LPI can be used at a sensitivity level of 2 or above (generally corresponding to an Ra below 3.2 pm). This ability to use an LPI dye at a sensitivity level of 2 or above is critical for quality control of components produced for the aerospace industry.

Another significant advantage of the method is that hydrogen ingression at the article surface is negligible. The present method can therefore be used to smooth the surface of such articles without compromising the strength or fatigue properties of the article surface.

The method of the present invention is such that the improvement of the surface smoothness can occur without deforming the surface such as by rounding of edges or geometric features. This is because the etching solution dissolves material from all the surfaces it makes contact with in a controlled, uniform and linear fashion. In particular, an important advantage of the present invention is that detail (such as lettering) and surface sharpness can be retained. This is surprising, as it would be expected that the reaction conditions, which are necessarily aggressive, remove detail and lead to rounding of edges.

Many chemical etching methods make use of hydrofluoric acid, which is known to be a hazardous chemical to stockpile and handle. The present invention provides a method that does not require the use of hydrofluoric acid. The inventors have surprisingly found that exceptional levels of smoothness can be obtained using ammonium bifluoride. This provides for a significant improvement in the safety profile of the method as compared with methods that use hydrofluoric acid. Preferably the etching solution is not made by using hydrofluoric acid, sulfuric acid or hydrochloric acid.

As a further point, the materials used in the method and the products of the method are environmentally benign to the extent that they can be disposed of in the conventional waste stream, without needing specialised and expensive processing. This ensures that the method remains as economically viable as possible from a waste disposal perspective.

The conditions set out below provide for further improvements to the above properties.

Regarding the nitric acid, the etching solution can comprise 25-40 v/v% nitric acid, preferably 28-38 v/v% nitric acid, more preferably about 30-35 v/v% nitric acid. As would be evident to the skilled person, nitric acid is supplied commercially as a solution in water. The nitric acid in the present invention is a 68 wt% solution of nitric acid in water, as is most readily commercially available.

Regarding the ammonium bifluoride, the etching solution can comprise 5-12 wt% ammonium bifluoride, preferably 6-10 wt% ammonium bifluoride, more preferably about 7-9 wt% ammonium bifluoride.

The etching solution is aqueous, with the nitric acid and ammonium bifluoride being mixed with/dissolved in water. Optionally other ingredients can be present. Preferably the etching solution consists of nitric acid, ammonium bifluoride and water.Regarding the temperature, the etching solution can be maintained at a temperature between 5 and 60 °C, preferably between 10 and 40 °C, more preferably between 15 and 30 °C.

When the article is the product of electron beam additive manufacturing it can be immersed in the etching solution for between 40 and 160 minutes, preferably between 60 and 140 minutes, more preferably between 80 and 120 minutes. When the article is the product of laser beam additive manufacturing it can be immersed in the etching solution for between 10 and 120 minutes, preferably between 30 and 100 minutes, more preferably between 50 and 80 minutes.

When the article is the product of electron beam additive manufacturing it can be immersed in the etching solution until between 500 and 2000 pm of surface material of the article has been removed, preferably between 800 and 1700 pm, more preferably between 1000 and 1500 pm. When the article is the product of laser beam additive manufacturing it can be immersed in the etching solution until between 125 and 750 pm of surface material of the article has been removed, preferably between 200 and 600 pm, more preferably between 300 and 500 pm.

The article can be immersed in the etching solution until the surface roughness has reached 8 Ra or lower, preferably 5 Ra or lower, more preferably 3 Ra or lower. Particularly preferred surface roughness points are below 5 Ra or below 3.2 Ra. The article can immersed in the etching solution until the surface roughness has reached 40 Rt or lower, preferably 30 Rt or lower, more preferably 20 Rt or lower. The article can be immersed in the etching solution until the surface roughness has reached 30 Rz or lower, preferably 15 Rz or lower, more preferably 10 Rz or lower. Monitoring Ra, Rt and/or Rz parameters provide for simple ways to determine the length of time that an article must be immersed in the etching solution for to produce the required level of smoothness.

The etching solution can flow relative to the article at a flow rate between 0.1 and 5.0 cm/s over the article. The flow is preferably a laminar flow prior to contacting the article. The flow rate can be between 0.05 and 3 cm/s over the article, preferably between 0.1 and 2 cm/s over the article, more preferably between 0.5 and 1.5 cm/s over the article. These flow rates allow for the etching solution to reach all surfaces of the article but without encouraging significant turbulence. Turbulent flow can result in uncontrolled and varied material removal rates, leading to uneven surfaces, rounding of edges and deformation of surface geometries.

The etching solution can flow relative to the article at a flow rate that maintains the surface temperature of the article within lower and upper boundaries, wherein the lower boundary is the bulk etchant solution temperature + 0.5°C and the upper boundary is 10°C. In preferred embodiments, the lower boundary is the bulk etchant solution temperature + 1°C and the upper boundary is 8°C, preferably the lower boundary is the bulk etchant solution temperature + 1.5°C and the upper boundary is 7°C, more preferably the lower boundary is the bulk etchant solution temperature + 2°C and the upper boundary is 6°C.

The etching solution can flow relative to the article at a flow rate that maintains at least one edge of the article at a radius of curvature less than 15 pm, preferably less than 10 pm, more preferably less than 5 pm.

According to a second aspect, the invention provides a surface treated additive manufactured titanium or titanium alloy article produced according to the method of the first aspect of the invention.

Brief Description of the Drawings

Embodiments of the invention will now be described by way of example only with reference to the accompanying figures in which:

Figure 1 shows a process flow chart of a preferred embodiment of the method of the invention;

Figure 2 shows photographs of a 3D printed titanium article before and after chemical etching according to the invention;

Figure 3 shows photographs of a 3D printed titanium article before and after chemical etching according to the invention, wherein surface roughness has been visualised using liquid penetrate inspection;

Figure 4 shows photographs of a 3D printed titanium article before and after chemical etching according to the invention, wherein surface roughness has been visualised using liquid penetrate inspection;

Figure 5 shows SEM (Scanning Electron Microscope) images of the surface of a 3D printed titanium article before and after chemical etching according to the invention;

Figure 6 shows microscope images of a hole feature in a 3D printed titanium article before and after chemical etching according to the invention;

Figure 7 shows an electron back-scatter diffraction image of the surface of a 3D printed titanium article after chemical etching according to the invention and shows only a hexagonal close phase alpha titanium phase and the absence of any titanium hydride phase;

Figure 8 shows a plot of stress v number of cycles;

Figure 9 shows a plot of change in Ra of an electron beam additive manufactured article with respect to immersion time; and

Figure 10 shows a plot of change in Ra of the article of Figure 8 with respect to amount of material removed.

Description

As set out above, a method is provided for increasing the surface smoothness of an additive manufactured titanium or titanium alloy article. Examples of titanium alloys that can be used with the method of the invention include any titanium alloys that are compatible with 3D printing. These include titanium alloys such as Ti6Al-4V (also known as Ti 6-4) e.g. Titanium 6A1-4V grade 5, Titanium 6A1-4V grade 23 ELI [Extra Low Interstitial] and Titanium grade 2 CP [Commercially Pure], Ti6A12Sn4Zr2Mo and TiA16Nb7.

The surface smoothness is improved by the removal of material. When a thickness of 15 pm or less is removed, it is generally referred to as chemical pickling. When a thickness of more than 15 pm is removed, it is generally referred to as chemical milling. The method according to the invention will typically be used to remove a thickness of more than 15 pm, and the method can therefore be referred to as a chemical milling method.

Chemical milling of titanium and titanium alloys can be classified into three classes: integral milling, graded milling and partial milling. Integral milling relates to milling of all surfaces to reduce thickness or remove contaminants. Graded milling is used to obtain an evolutive thickness of the parts. Partial milling involves the use of masking material to reduce thickness only of select parts.

The etching solution is an aqueous solution comprises 20-45 v/v% nitric acid (68 wt% nitric acid in water) and 4-14 wt% ammonium bifluoride. By this, we mean that the concentrations should be within this range prior to immersion of the article. The starting concentrations of the acids will decline with treatment time. Preferably, the concentrations remain within the lower boundaries until the treatment is complete. The volume in volume (v/v) and weight (wt) percentages are based on the total etching solution.

The solution is aqueous, and preferably consists of nitric acid and ammonium bifluoride with the balance being water. However, it can contain other ingredients as set out above.

The etching solution is maintained at a temperature between 2 and 85 °C. The method typically allows for the etching reaction to run effectively at room temperature, and the etching solution can simply be allowed to equilibrate with room temperature. This keeps the cost of the process low. The etching solution heats up as the process runs due to the exothermic etching reaction. Generally, the heating is not sufficient to go beyond the upper temperature boundary. However, active heating may be required if the incoming water is too cold. Active cooling can be employed if temperatures are expected to rise too much, which is a consideration with high surface area articles.

The article is immersed in an etching solution where the etching solution flows relative to the article. Typically, the article will be fixed in place and the etching solution will be in motion. Alternatively, the etching solution is stationary but the article is moved to generate flow around the article. It is important to keep the etching solution moving across the surfaces of the article because this prevents localised overheating events that can lead to overreaction and pitting. The method can be used for integral milling, i.e. the etchant solution can reach and etch all surfaces that a liquid can reach. Of course, if it is preferred that any surfaces are not etched, these surfaces can be masked prior to immersion in the etching solution. Such partial milling and masking techniques are well known in the art, and typically involve use of a resin that is impervious to the etchant solution. Graded milling can be achieved by successive cuttings of a mask material, or by slowly immersing or withdrawing the article from the etching solution.

When the article is the product of electron beam additive manufacturing it is preferably immersed in the etching solution for between 40 and 160 minutes. When the article is the product of laser beam additive manufacturing it is preferably immersed in the etching solution for between 10 and 120 minutes. These immersion timings dictate the extent of material removed from the 3D printed article. The extent of smoothing can be increased by removing larger amounts of material in the etching process. Of course, having to remove larger amounts of material has several drawbacks. More material must be used in the first place, which can be an expensive waste of material. There is also greater potential for loss of detail and edge rounding. There are also greater constraints on design parameters, as all surfaces would need to accommodate the larger levels of material removal. The process of the invention allows surface smoothing to occur with a minimum of material removal, and therefore minimises these drawbacks.

When the article is the product of electron beam additive manufacturing it is advantageously immersed in the etching solution until between 500 and 2000 pm of surface material of the article has been removed. When the article is the product of laser beam additive manufacturing it is advantageously immersed in the etching solution until between 125 and 750 pm of surface material of the article has been removed. It is straightforward to monitor how much material is removed from a surface. For example, a selected width of the article can be measured prior to immersion and then measured again at a set time after immersion. The difference relates to two opposing surfaces, and is divided by two to give the amount of material removed at a single surface. Immersion can be continued until the desired amount of surface material is removed. A number of data points at different times can be collected to extrapolate and predict when the desired amount of surface material is removed.

The article can be immersed in the etching solution until the surface roughness has reached 8 Ra or lower, preferably 5 Ra or lower, more preferably 3 Ra or lower. The article can be immersed in the etching solution until the surface roughness has reached 40 Rt or lower, preferably 30 Rt or lower, more preferably 20 Rt or lower. The article can be immersed in the etching solution until the surface roughness has reached 30 Rz or lower, preferably 15 Rz or lower, more preferably 10 Rz or lower.

Ra, Rt and Rz values are well known in the field and can be easily determined by the skilled person using the techniques described in EN ISO 4287. In more detail, Ra is the arithmetic average of the absolute values of the roughness profile ordinates. The average roughness is the area between the roughness profile and its mean line, or the integral of the absolute value of the roughness profile height over the evaluation length. Rt is the total height of the roughness profile. This is the difference between the height of the highest peak and depth of the deepest valley within the evaluation length. Rz is the average maximum peak to valley of five consecutive sampling lengths within the measuring length. Each of these values can be measured at given times after immersion to determine when the desired values have been achieved.

The etching solution can flow relative to the article at a flow rate between 0.1 and 5.0 cm/s over the article. The flow rate can be between 0.05 and 3 cm/s over the articfe, preferably between 0.1 and 2 cm/s over the article, more preferably between 0.5 and 1.5 cm/s over the article. The flow rate can be measured readily by a flow meter that has been adapted for compatibility with the etching solution.

The etching solution can flow relative to the article at a flow rate that maintains the surface temperature of the article within lower and upper boundaries, wherein the lower boundary is the bulk etchant solution temperature + 0.5°C and the upper boundary is 10°C. In a preferred embodiment the lower boundary is the bulk etchant solution temperature + 1°C and the upper boundary is 8°C, preferably the lower boundary is the bulk etchant solution temperature + 1.5°C and the upper boundary is 7°C, more preferably the lower boundary is the bulk etchant solution temperature + 2°C and the upper boundary is 6°C. The bulk etchant solution temperature can be measured readily by thermometer or other temperature probe. The surface temperature of the article can also be measured readily by thermometer or other temperature probe, such as an infrared temperature probe or optical fibre sensor.

The etching solution can flow relative to the article at a flow rate that maintains at least one edge of the article at a radius of curvature less than 15 pm, preferably less than 10 pm, more preferably less than 5 pm. The radius of curvature can be readily measured using known techniques.

The invention also provides a surface treated additive manufactured titanium or titanium alloy article produced according to the method of the first aspect of the invention. The article is preferably for use in the aerospace or medical industries. For example, thermal, aerodynamic and structural efficiency can be improved for typical applications such as the article being piping, ducting, brackets and exchangers. Also Radio Frequency [RF] for wave guides. The article could be a medical implant such as hip and knee joints. Space is also seen as a growing sector of interest for this invention.

Examples

Figure 1 shows a process flow chart of a preferred embodiment of the invention. A 3D printed titanium or titanium alloy article is first subjected to an alkaline clean using a solution comprising sodium tetraborate decahydrate and fatty alcohol ethoxylate. After rinsing with deionised water, the article is subjected to the etching solution (referred to as the Almbrite solution in Figure 1). After etching, the article is again rinsed with deionised water, neutralised with sodium carbonate, and then subjected to a final rinse. The article is then dried. If the article is particularly rough, it may need to be sandblasted prior to the alkaline clean.

In a typical process the article is manufactured from Ti6A14V using the laser beam powder bed process. It fits within a volume of approximately 75 x 75 x 100mm. The solution was made up of 33% Nitric acid by volume and 7.9 % of ammonium bifluoride by weight. In more detail, the nitric acid solution (68 wt% nitric acid in water) is added to a tank. Water is then added and mixed and then the ammonium biflouride is added and mixed, to the stated concentrations. Solution temperature was set at 18°C with a processing time of 120 minutes. The volume of the solution used was 450L. The article was in a fixed position and solution movement was pumped around the article at a rate of 5cm/sec.

Figure 2 shows a 3D printed titanium article before (left image) and after (right image) treatment according to the invention. It can be seen that before treatment the article has a rough surface, measuring about 18 Ra. After treatment, the surface is smooth enough to have a shiny appearance, measuring below 3.2 Ra. Where the edges were sharp in the original article, the sharpness of the edges has been retained.

Figure 3 shows the article of Figure 2 again before (left image) and after (right image) undergoing the treatment according to the invention. In Figure 3, the surface smoothness has been visualised by liquid penetrant inspection (LPI) with sensitivity level 2. The bright colouring is fluorescence, showing where the liquid penetrant has been retained. In the left image, a significant amount of dye has been retained by the roughness of the surface. In contrast, in the right image very little dye has been retained, with a significant proportion of the surface retaining no dye. This demonstrates how smooth the surface has become and also how articles processed by the method of the invention become compatible with the use of LPI in detecting surface cracks and other defects.

Figure 4 shows a disc of titanium that has been 3D printed, both before (left image) and after (right image) treatment according to the invention. As with Figure 3, in Figure 4 the surface smoothness has been visualised by LPI.

Figure 5 shows microscope images of a 3D printed titanium article before (upper image) and after (lower image) being subjected to the treatment of the invention. The upper image shows the surface roughness in detail, comprising a large number of unfused or partially fused titanium particles. After the treatment, shown in the lower image, the surface shows no such particles.

Figure 6 shows microscope images of a hole feature in a 3D printed titanium article before (upper image) and after (lower image) being subjected to the treatment of the invention. As well as seeing smoothing of the surface, the rough edge has become sharp and well defined.

Figure 7 shows an electron back-scatter diffraction image of the surface of a 3D printed titanium article after chemical etching according to the invention and shows only a hexagonal close phase alpha titanium phase and the absence of any titanium hydride phase.

Example 1

A series of 8 titanium articles were prepared by an identical laser additive manufacture process. Each was subjected to the same chemical etching process.

The articles were cylindrical test specimens, approximately 90mm in length and 20 mm in diameter. The article is manufactured from Ti6A14V using the laser beam powder bed process. The solution was made up of 33% by volume nitric acid and 7.5 % by weight of ammonium bifluoride with the remainder being water. In more detail, the nitric acid solution (68 wt% nitric acid in water) is added to a tank. Water is then added and mixed and then the ammonium biflouride is added and mixed, to the stated concentrations. Solution temperature was set at 17°C with a processing time of 120 minutes. The volume of the solution used was 450L. The articles were in a fixed position and solution movement was pumped around the article at a rate of 5cm/sec.

Ra, Rt and material removal were measured at 45 minutes and 57 minutes. The results are shown in Table 1, below.

From the measurement at 45 minutes, it was possible to extrapolate that the desired Ra of 3.2 Ra or below should be reached for all samples at 57 minutes.

Table 1

Ref	Time = 0 min		Time = 45 min
	Starting diameter (mm)	Ra	Rt		Material removed per surface (pm)	Material removal rate (p/min)	Ra	Rt
1	12.81	14.6	114.1		487	10.8	4.9	33.6
2	12.772	12.7	105.3		488	10.8	3.3	23.7
3	12.764	12	99.6		461	10.2	3	19.1
4	12.735	16.1	153.4		481	10.7	4.1	28
5	12.758	11.3	94.9		484	10.8	2.8	17.9
6	12.772	14.2	108		471	10.5	3.5	23.6
7	12.778	10.6	90.3		497	11.0	4	25.1

8	12.754	12.3	87.1		481	10.7	3.6	21.6
Ave	12.768	13	106.6		481	10.7	3.7	24.1
					Time = 57 min
1					581	10.2	3.2	24.1
2					581	10.2	3	21.6
3					578	10.1	2.5	19.6
4					584	10.2	2.6	22.1
5					588	10.3	2.8	21.1
6					605	10.6	2.8	26.6
7					586	10.3	3.1	19.5
8					591	10.4	2.4	15.4
Ave					587	10.3	2.8	21.3

The articles were also analysed to demonstrate the effect of the chemical etching process of the invention on fatigue performance. Figure 8 is a fatigue graph. It shows an S-N curve, also known as a Stress versus number of cycles to failure curve, which is the standard method to present fatigue performance. As the stress increases the number of cycles to failure decreases thus a curve is created.

As shown on this fatigue graph, treating the specimens to a surface roughness of 3.2 Ra has halved the difference in fatigue strength between a fully machined and untreated specimen. It is expected that as the surface roughness reduces the fatigue strength will continue to increase due to the reduction in surface crack propagation areas.

Example 2

A titanium article was prepared by electron beam additive manufacture. The article was subjected to chemical etching with 32% by volume nitric acid and 7.7 % by weight of ammonium bifluoride @ 23°C for 147 minutes achieving a 1.6Ra surface finish. In more detail, the nitric acid solution (68 wt% nitric acid in water) is added to a tank. Water is then added and mixed and then the ammonium biflouride is added and mixed, to the stated concentrations.Figure 9 shows a plot of Ra versus immersion time. Figure 10 shows a plot of Ra versus material removal. It can be seen that articles prepared by electron beam additive manufacture have a higher starting Ra than articles prepared by laser additive manufacture. It can also be seen that Ra improves in a largely predictable manner that allows for extrapolation to determine when the desired Ra should be reached.

Example 3

A titanium article was prepared by Laser Beam powder bed process. Processing according to the invention was carried out with 33% by volume nitric acid and 8.8 % of ammonium bifluoride. The etching solution was made as in Example 1. Solution 5 temperature at 21 °C for 120 minutes. The article was fixed with the solution moving around the article at a rate of 5cm/sec. Hydrogen measurements were conducted on samples taken from close to the surface of the articles, and the results are shown below. Vacuum hot extraction was the method used to measure the amount of hydrogen found in the sample.

Table 2

Specimen	Hydrogen (ppm)
	Repeat 1	Repeat 2	Repeat 3
As manufactured	19	14	13
Etched	16	16	19
Etched and heat treated	31	24	22

Hydrogen contents of above 150ppm generally render an article too brittle to be of use. Typically, it is desirable to achieve as low a hydrogen content as possible. The table above shows that the etching conditions according to the present invention do 15 not increase the hydrogen content at the surface by any meaningful amount. The invention therefore provides for a way to prevent the significant quantities of hydrogen generated at the surface from becoming ingressed in the surface.

Claims (16)

Claims

1. A method of increasing the surface smoothness of an additive manufactured titanium or titanium alloy article, the method comprising immersing the article in an etching solution, wherein the etching solution flows relative to the article, and wherein the etching solution comprises 20-45 v/v% nitric acid and 4-14 wt% ammonium bifluoride, and the etching solution is maintained at a temperature between 2 and 85 °C, wherein the article is the product of electron beam additive manufacturing and is immersed in the etching solution for between 40 and 160 minutes, or wherein the article is the product of laser beam additive manufacturing and is immersed in the etching solution for between 10 and 120 minutes.

2. A method according to claim 1, wherein the etching solution comprises 25-40 v/v% nitric acid, preferably 28-38 v/v% nitric acid, more preferably about 30-35 v/v% nitric acid.

3. A method according to any preceding claim, wherein the etching solution comprises 5-12 wt% ammonium bifluoride, preferably 6-10 wt% ammonium bifluoride, more preferably about 7-9 wt% ammonium bifluoride.

4. A method according to any preceding claim wherein the etching solution is not made using sulphuric acid, hydrochloric acid or hydrofluoric acid.

5. A method according to any preceding claim, wherein the etching solution is maintained at a temperature between 5 and 60 °C, preferably between 10 and 40 °C, more preferably between 15 and 30 °C.

6. A method according to any preceding claim, wherein the article is the product of electron beam additive manufacturing and is immersed in the etching solution until between 500 and 2000 pm of surface material of the article has been removed, or wherein the article is the product of laser beam additive manufacturing and is immersed in the etching solution until between 125 and 750 pm of surface material of the article has been removed.

7. A method according to any preceding claim, wherein the article is immersed in the etching solution until the surface roughness has reached 8 Ra or lower, preferably 5 Ra or lower, more preferably 3 Ra or lower.

8. A method according to any preceding claim, wherein the article is immersed in the etching solution until the surface roughness has reached 40 Rt or lower, preferably 30 Rt or lower, more preferably 20 Rt or lower.

9. A method according to any preceding claim, wherein the article is immersed in the etching solution until the surface roughness has reached 30 Rz or lower, preferably 15 Rz or lower, more preferably 10 Rz or lower.

10. A method according to any preceding claim, wherein the etching solution flows relative to the article at a flow rate between 0.1 and 5.0 cm/s over the article.

11. A method according to claim 10, wherein the flow rate is between 0.05 and 3 cm/s over the article, preferably between 0.1 and 2 cm/s over the article, more preferably between 0.5 and 1.5 cm/s over the article.

12. A method according to any preceding claim, wherein the etching solution flows relative to the article at a flow rate that maintains the surface temperature of the article within lower and upper boundaries, wherein the lower boundary is the bulk etchant solution temperature + 0.5°C and the upper boundary is 10°C.

13. A method according to claim 12, wherein the lower boundary is the bulk etchant solution temperature + 1°C and the upper boundary is 8°C, preferably wherein the lower boundary is the bulk etchant solution temperature + 1.5°C and the upper boundary is 7°C, more preferably wherein the lower boundary is the bulk etchant solution temperature + 2°C and the upper boundary is 6°C.

14. A method according to any preceding claim, wherein the etching solution flows relative to the article at a flow rate that maintains at least one edge of the article at a radius of curvature less than 15 pm, preferably less than 10 pm, more preferably less than 5 pm.

15. A surface treated additive manufactured titanium or titanium alloy article produced according to the method of any one of claims 1 to 14.

16. An article according to claim 15, for use in the aerospace or medical industry.