The present invention relates to a sputtering target comprising a NiSi alloy which comprises from 2 to 8 weight% Si.

The use of NiSi alloys as sputtering targets is known, for example, for the deposition of under-bump metallization layers, barrier layers for soldering processes, any high Ni containing layers like blocker layers in low-e stacks to protect the Ag-layer underneath form oxidation, metal mesh electrodes in touch panels and also as replacement for any other high-Ni containing nonmagnetic layer.

In particular, it is known from US 2008/0138974 that NiSi alloys with a content of 2 to 20 weight% of Si can be used as sputtering targets to deposit barrier layer films for under-bump metallization.

Furthermore, US 6,780,295 discloses about melting, casting and rolling a NiSi >4.5-50wt% alloy in order to obtain a sputtering target with at least 90% magnetic pass through flux compared to a nonmagnetic metal.

However, the known NiSi sputtering targets could still be improved as regards their sputtering properties and the properties of the sputtered films.

Thus, it is the object of the present invention to provide a NiSi alloy based sputtering target which allows for sputtering with a high reproducibility of the sputtering rate and under high power without abnormal discharge (arcing). Furthermore, the layers/films obtained by sputtering using said target should show improved uniformity.

These objects are achieved by a sputtering target comprising, or consisting of, a NiSi alloy comprising from 2 to 8 weight% Si, having an average grain size of 150 to 400 pm and a grain size variation of lower than 30%.

It has been discovered by the inventors that sputtering using the target of the invention can be performed with high power of up to 20 W/cm ² and that the deposited films are characterized by a high film uniformity. The sputtering target of the invention comprises a sputtering material which comprises the NiSi alloy in any one of the embodiments as described herein. Preferably, the NiSi alloy represents at least 99 wt% of the sputtering material and more preferably, the sputtering material consists of the NiSi alloy, the remainder being unavoidable impurities.

Preferably, in the sputtering target according to the invention the NiSi alloy comprises from 2.5 to 7.5 weight% Si, more preferably comprises from 2.75 to 7 weight% Si, and most preferably comprises from 3 to 6 weight% Si.

In further embodiments, the NiSi alloy preferably consists of Ni and from 2 to 8 weight% Si, more preferably consists of Ni and from 2.5 to 7.5 weight% Si, still more preferably consists of Ni and from 2.75 to 7 weight% Si, and most preferably consists of Ni and from 3 to 6 weight% Si with the remainder being unavoidable impurities.

Optionally, in the sputtering target according to the invention the NiSi alloy further comprises one or more other metal elements selected from V, Nb, Ti, Ir, Pd or Mn in a total amount of up to 8 wt%. Preferably, in the sputtering target according to the invention the NiSi alloy further comprises one or more other metal elements selected from V, Nb, Ti, Ir, Pd or Mn in a total amount of from 0.1 to 8 wt%, preferably from 1 to 6 wt%, more preferably from 2.5 to 5.5 wt%, and most preferably from 2 to 5 wt%. More preferably, the other metals are selected from V, Nb, or Ti, most preferred the other metal is V.

In further embodiments the NiSi alloy preferably consists of Ni, from 2 to 8 weight%, preferably from 2.5 to 7.5 wt.%, more preferably from 2.75 to 7 wt.%, still more preferably 3 to 6 wt.%, Si and, optionally, one or more other metal elements selected from V, Nb, Ti, Ir, Pd or Mn in a total amount of up to 8 wt.%, preferably from 0.1 to 8 wt, more preferably from 1 to 6 wt%, still more preferably from 2.5 to 5.5 wt%, and most preferably from 2 to 5 wt%, with the remainder being unavoidable impurities.

More preferably, the other metals are selected from V, Nb, or Ti, most preferred the other metal is V. The presence of an other metal element selected from V, Nb, Ti, Ir, Pd or Mn, preferably V, Nb or Ti in an amount as recited facilitates an alloy having beneficial average grain sizes and a grain size variation.

Accordingly, the NiSi alloy preferably comprises from 84 to 98 wt% Ni, more preferably comprises from 87 to 96.5 wt% Ni, and most preferably comprises from 90 to 95 weight% Ni.

Optionally, the Si used for producing the alloy may contain dopants such as B, P or As in the usual amounts. The dopant(s) will then also be present in the alloy used in the sputtering target.

Thus, the NiSi alloy which preferably consists of Ni and from 2 to 8 weight% Si, more preferably consists of Ni and from 2.5 to 7.5 weight% Si, still more preferably consists of Ni and from 2.75 to 7 weight% Si, and most preferably consists of Ni and from 3 to 6 weight% Si with the remainder being unavoidable impurities and the NiSi alloy in any other embodiments described herein, optionally may contain dopant elements.

For sake of clarity the term“dopant elements” designates elements which have been deliberately introduced into the Si in a defined amount, in contrast to (unavoidable) impurities.

The term“usual dopant amounts” as used herein designates amounts which preferably do not exceed 1 wt.%, more preferably do not exceed 0.5 wt.%, for each dopant element in the Si used for preparing the alloy, and more preferably designate amounts in which the total amount of all dopant elements does not exceed 1 wt.%, more preferably does not exceed 0.5 wt.%, in the Si used for preparing the alloy.

In the finally prepared NiSi alloy the optionally present dopant elements are each preferably present in an amount of 1000 ppm or less, more preferably of 500 ppm or less, and most preferably of 100 ppm or less, and more preferably the optionally present dopant elements are present in a total amount of 1000 ppm or less, more preferably of 500 ppm or less, and most preferably of 100 ppm or less. The average grain size of the NiSi alloy is preferably from 175 to 375 pm, and more preferably from 200 to 350 pm.

The NiSi alloy preferably is produced by melting Ni and Si. Preferably, Ni with a purity of at least 99.9 % (3N), more preferably of at least 99.95 (3.5N), more preferably at least 99.99 % (4N), more preferably of 99.995 % (4.5N), and most preferably of at least 99.999 % (5N) is used for producing the alloy.

Further preferred, the Si used in producing the alloy has a purity of at least 99.99 % (4N), more preferably of 99.995 % (4.5N), and most preferably of at least 99.999 % (5N), excluding dopants.

Further preferred, if present, the V, Nb, Ti, Ir, Pd and/or Mn used in producing the alloy has a purity of at least 99.95 (3.5N), more preferably at least 99.99 % (4N), still more preferably of 99.995 % (4.5N), and most preferably of at least 99.999 % (5N).

As mentioned above, the Si feedstock may be doped with dopants such as B, P or As as defined above.

The present invention furthermore relates to a process for forming a sputtering target, preferably a sputtering target in any of the embodiments as herein described, which comprises rolling an NiSi alloy comprising 2 to 8 weight% Si and optionally other metal elements in a total amount of up to 8 wt% selected from V, Nb, Ti, Ir, Pd or Mn, or a NiSi alloy in any other embodiment as described herein, and annealing it at a temperature in the range of 750 to 975 °C.

In particular the annealing step at the prescribed temperature is important to obtain the grain sizes and grain size variation yielding the improved sputtering properties.

Rolling preferably takes place at a temperature of 1 ,000 to 1 ,250 °C.

Rolling is effected in one pass or multiple, i.e. two or more, passes, preferably multiple passes. When using multiple rolling passes the ingot may have to be re-heated between the different passes.

It is preferred that rolling is effected so that a total thickness reduction, i.e. a thickness reduction after all rolling passes have been completed, of the ingot of 60 to 90 % is obtained.

Annealing is performed at a temperature of from 750 to 975 °C, preferably at a temperature of from 800 to 950 °C, and more preferably from 825 to 940 °C.

Preferably, the annealing time is from 25 minutes to 4 hours, more preferably is from 35 minutes to 3 hours, and most preferably is from 45 minutes to 2 hours. Usually an annealing time of 1 hour will be applied.

Usually, the rolling and annealing steps are preceded by a step in which the Ni and Si are melted/cast together to form an ingot. This may be done by overheating the melt by 150 to 300 °C for casting under inert gas or in vacuum.

Definitions/Measurement methods a) Grain size and grain size variation

In order to determine the grain size variation, a sputtering target of a diameter of at least 150 mm is prepared. The target is divided into 4 equal sectors by drawing 2 lines under 90° angle through the center of the plate.

Coin shaped samples with d = 1 to 2cm are taken close to the edge of the plate at each of the 4 positions where the lines exit the plate, as well as at the center of the plate where the lines intersect. These 5 samples are ground starting with 120 grit down to 2400 grit SiC based grinding paper and subsequently etched by H2O - HNO3 - HCI - H2O2 (ca. 20ml - 15ml - 20ml - 10ml) on one side for about 1 -2 minutes at room temperature, until a visible grain structure develops.

After evaluation of the grains sizes (see below) the procedure is repeated for the opposite side of each sample. The grain sizes are evaluated as follows:

From each micrograph a photo is taken. The magnifications is set in a way that a line drawn over a photograph of the etched samples crosses between 8 and 30 grain boundaries.

Each micrograph is evaluated by the line intersection method (ASTM E 1 12) and the mean grain size is calculated for each of the 10 samples. In addition the largest grain found by the method is noted for each of the 10 samples.

The average grain size M was determined by the linear intercept method according to DIN EN ISO 643 according to the following equation:

M=(L ^* p)/(N ^* m) wherein

L: length of the measurement line p: number of measurement line N: number of the cut objects m: magnification

The values were determined at 3 ^* 3=9 different measurement locations, each in three depths: 0 mm, 3 mm and 6 mm.

From the so determined grain sizes M the grain size variation can be calculated according to the following equations (as A1 value or, alternatively, as B1 value):

A1 =( Mmax ^— Mave)/(Mave * 1 00)

B1 = (Mave ^— Mmin)/( Mave * 1 00) with

Mmax: maximum grain size

Mmin : minimum grain size Mave : average grain size

The grain size variation as herein used is defined as the higher value of the variation calculable from A1 and B1 . b) Sputtering performance Sputtering performance was evaluated on a ZH-1000-3K sputtering machine from Onlink Technologies, Germany.

D75mmx6mm targets were machined from the rolled plates. As power supply an Advanced Energy Pinnacle ⁺ was used.

Sputtering deposition as done as follows: Process pressure: 3 x 10 ³ mbar Ar-flow: 200sccm Power: Burn-in for 0.5h at 2W/cm ² . Afterwards continuous increase up to 20 W/cm ² .

Arcing was monitored on the Advanced Energy Pinnacle ⁺ power supply with the following settings:

- Arc shutdown = 200ps

- Arc trip level = 70V

Examples Several examples and comparative examples were performed in order to illustrate the invention. All target plates were manufactured by using Ni feedstock of 3N-5N purity, Si of 5N purity, V of 5N, Nb of 5N and Mn of N5. For examples 5 to 7 the target plates we manufactured by using a Ni feedstock of 3N5 purity. All castings were made with 30 kg of ingot material under vacuum or inert gas in an induction furnace. Casting was done by superheating the melt by 150 to 300°C into a steel or graphite mold.

As-cast billets were rolled and annealed as described below. Rolling reduction was from ingot thickness of 75 mm down to a plate thickness of 12 mm (corresponding to a thickness reduction of 84%). For annealing, a preheated air furnace was used with 1 h soaking time, followed by air cooling.

After annealing, the plates were machined to a thickness of 6 mm to obtain the sputtering targets.

The obtained results are listed in Table 1 below: Table 1 :