TWI882827B - Tin-based alloy and tin ball - Google Patents

Abstract

A tin-based alloy comprises, based on 100 wt% of the total amount of the tin-based alloy, 2 wt% to 5 wt% of silver, 0.1 wt% to 2 wt% of bismuth, 0.02 wt% to 0.07 wt% of nickel, 0.002 wt% to 0.006 wt% of phosphorus, and the remainder of tin; and the tin-based alloy does not contain lead, copper, and germanium. The present invention also provides a tin ball formed of the tin-based alloy. The tin-based alloy of the present invention does not contain lead, copper and germanium. Through the coordination of 2wt% to 5wt% of silver, 0.1wt% to 2wt% of bismuth, 0.02wt% to 0.07wt% of nickel, 0.002wt% to 0.006wt% of phosphorus and the remainder of tin, the tin ball of the present invention formed by the tin-based alloy has good oxidation resistance. The solder joint formed by the tin ball after multiple reflows has only one intermetallic layer, so that the solder joint is not prone to electrical connection failure. The solder joint has good or even excellent interface fracture morphology and maintains a certain level of thermal fatigue resistance, so that the solder joint has high reliability.

Description

Tin-based alloy and tin ball

The present invention relates to an alloy for soldering, in particular to a tin-based alloy that does not contain lead, copper and germanium, and a tin ball formed from the tin-based alloy.

Taiwan Patent Publication No. 789165 discloses a lead-free copper-free tin alloy. Based on the total weight of the lead-free copper-free tin alloy being 100wt%, the lead-free copper-free tin alloy comprises 0.01wt% to 3.0wt% of silver (Ag), 2.0wt% to 3.0wt% of bismuth (Bi), 0wt% to 2.0wt% of antimony (Sb), 0.005wt% to 0.1wt% of nickel (Ni), 0.005wt% to 0.02wt% of germanium (Ge), and the remainder of tin (Sn). Because the lead-free copper-free tin alloy contains the above-mentioned components and the corresponding weight percentage ranges, the solder bump (also called solder joint) formed by welding the lead-free copper-free tin alloy has good mechanical impact reliability, weldability, ductility, oxidation resistance and thermal cycle reliability (i.e. thermal fatigue resistance).

Therefore, the first object of the present invention is to provide a tin-based alloy that is different from the lead-free and copper-free tin alloy and has a novel composition.

Therefore, the tin-based alloy of the present invention comprises, based on 100wt% of the total amount of the tin-based alloy, 2wt% to 5wt% of silver, 0.1wt% to 2wt% of bismuth, 0.02wt% to 0.07wt% of nickel, 0.002wt% to 0.006wt% of phosphorus, and the balance of tin; and the tin-based alloy does not contain lead, copper, and germanium.

The second object of the present invention is to provide a solder ball.

Therefore, the tin ball of the present invention is formed from the tin-based alloy as described above.

The efficacy of the present invention is that: the tin-based alloy of the present invention does not contain lead, copper and germanium, but through the cooperation of 2wt% to 5wt% of silver, 0.1wt% to 2wt% of bismuth, 0.02wt% to 0.07wt% of nickel, 0.002wt% to 0.006wt% of phosphorus and the balance of tin, the tin ball of the present invention formed by the tin-based alloy has good oxidation resistance; a solder joint formed by the tin ball after multiple reflows has only one intermetallic layer, so that the solder joint is not prone to electrical connection failure; the solder joint has a good or even excellent interface fracture morphology, and maintains a certain level of thermal fatigue resistance, so that the solder joint has high reliability.

The tin-based alloy of the present invention comprises, based on 100wt% of the total amount of the tin-based alloy, 2wt% to 5wt% of silver (Ag), 0.1wt% to 2wt% of bismuth (Bi), 0.02wt% to 0.07wt% of nickel (Ni), 0.002wt% to 0.006wt% of phosphorus (P), and the balance of tin (Sn); and the tin-based alloy does not contain lead (Pb), copper (Cu) and germanium (Ge).

The solder ball of the present invention is formed by the tin-based alloy, and the solder ball is suitable for soldering an electronic component to a solder pad (such as a nickel-gold substrate or a copper substrate) to form an electrically conductive solder joint between the electronic component and the solder pad. In some embodiments, the diameter of the solder ball ranges from, for example but not limited to, 0.05 mm to 1 mm, and is suitable for various integrated circuit packaging processes.

In the present invention, the solder ball is subjected to a high temperature baking test and a high temperature and high humidity test, thereby obtaining an evaluation result of "good, acceptable, and poor" for the oxidation resistance of the solder ball. In the present invention, the solder ball and the solder pad are reflowed multiple times (specifically three times) to form a solder joint, and then the solder joint is subjected to a thrust test and a temperature cycle test, thereby obtaining an evaluation result of "excellent, good, acceptable, and poor" for the interface fracture morphology and thermal fatigue resistance of the solder joint.

The following is a detailed description of each ingredient and content.

Based on 100wt% of the total amount of the tin-based alloy, the content of the silver is in the range of 2wt% to 5wt%, so that the solder joint formed by the tin-based alloy has good or even excellent soldering strength and interface fracture morphology, and thermal fatigue resistance above a fair level.

Based on 100wt% of the total amount of the tin-based alloy, the content of the bismuth is in the range of 0.1wt% to 2wt%, so that the solder joint formed by the tin-based alloy has good or even excellent hardness, welding strength and interface fracture morphology, and thermal fatigue resistance above a fair level. The tin-based alloy of the present invention includes bismuth in a content range of 0.1wt% to 2wt% but does not include copper, so that the melting point of the tin-based alloy of the present invention is close to the melting point (about 217°C) of the tin-silver-copper alloy commonly used in the industry (specifically SAC105~405), so that the tin-based alloy is suitable for the reflow curve of the tin-silver-copper alloy. The tin-based alloy does not need to re-adjust the reflow curve during use, so there is no production cost derived from adjusting the reflow curve. In some embodiments, the melting point of the tin-based alloy ranges from 210°C to 222°C.

Based on 100wt% of the total amount of the tin-based alloy, the nickel content ranges from 0.02wt% to 0.07wt%, so that the solder joint formed by the tin-based alloy has good or even excellent welding strength and interface fracture morphology, and thermal fatigue resistance above a fair level.

Taking the total amount of the tin-based alloy as 100wt%, the phosphorus content ranges from 0.002wt% to 0.006wt%, so that the tin ball formed by the tin-based alloy has a dense anti-oxidation layer to prevent the tin ball from being oxidized by external oxygen to form an oxide layer composed of SnO and _SnO2 , thereby allowing the tin ball to have good oxidation resistance (also known as anti-color change ability) in a high temperature environment.

In some embodiments, the tin-based alloy further comprises indium (In) in a content range of 0.2wt% to 1.5wt%. Taking the total amount of the tin-based alloy as 100wt%, the content of the indium is in a range of 0.2wt% to 1.5wt%, so that the solder joint formed by the tin-based alloy has better hardness, welding strength, interface fracture morphology and thermal fatigue resistance.

In the field of integrated circuit packaging, a solder ball made of the tin-silver-copper alloy is usually used to perform multiple reflows on a solder pad (such as a nickel-gold substrate or a copper substrate) to form a solder bump. When the solder pad is a nickel substrate, the intermetallic compounds in the solder bump include (CuNi) ₆ Sn ₅ and (NiCu) ₃ Sn ₄ , so there will be two intermetallic layers in the solder bump; when the solder pad is a copper substrate, the intermetallic compounds in the solder bump include Cu ₆ Sn ₅ and Cu ₃ Sn, so there will be two intermetallic layers in the solder bump. Since there are more than one intermetallic layer in the above two types of solder bumps, it is easy for one layer to be detached due to the Cu concentration effect, resulting in electrical connection failure of the solder bumps. The solder ball formed by the tin-based alloy of the present invention has only one intermetallic layer in a solder joint formed by multiple reflows on the solder pad, so that the solder joint is not prone to electrical connection failure. More specifically, for example, when the solder pad is a nickel-gold substrate, the solder joint formed has only one intermetallic layer formed by Ni ₃ Sn ₄ , as shown in FIG. 1 . For another example, when the pad is a copper substrate, a solder joint is formed with only one intermetallic layer formed of (CuNi) ₆ Sn ₅ , as shown in FIG. 2 .

In some embodiments, the tin-based alloy further comprises 0.2wt% to 1.5wt% of indium, and the solder ball formed by the tin-based alloy is reflowed on the solder pad multiple times to form a solder joint, which has only one fine-structured, thin-thickness, and non-brittle boundary metal layer, so the solder joint also has the advantage of being difficult to break, and thus the solder joint is less likely to have electrical connection failure. More specifically, for example, when the solder pad is a nickel-gold substrate, the solder joint formed has only one fine-structured, thin-thickness, and non-brittle boundary metal layer formed by Ni ₃ (SnIn) ₄ , as shown in FIG. 3 . For another example, when the pad is a copper substrate, a solder joint is formed with only one intermetallic layer formed of (CuNi) ₆ (SnIn) ₅ with a fine structure, a thin thickness and being not easily brittle, as shown in FIG. 4 .

The present invention will be further described with respect to the following embodiments, but it should be understood that the embodiments are only for illustration and description and should not be construed as limitations on the implementation of the present invention.

〈Preparation of Master Alloy〉

The preparation steps of tin-silver master alloy are as follows: (1). When the voltage of a frequency converter reaches 150V, pure tin (purchased from Timah, Indonesia, with a purity of 3N) and silver metal (purchased from South Korea) are added to the frequency converter; (2). After the feeding is completed, the voltage of the frequency converter is adjusted to 430V and the temperature is maintained for 5 minutes to completely melt the pure tin into liquid state, and to evenly disperse and melt the silver metal into the pure tin, thereby forming a tin-silver molten liquid; (3). Use a stirrer to stir the tin-silver melt for 5 seconds, then adjust the voltage of the oscillating furnace to 500V and keep the temperature for 2 minutes to make the pure tin and the silver metal in the tin-silver melt more evenly mixed to form a tin-silver mixed liquid; and (4). After adjusting the voltage of the oscillating furnace to 300V, cast the tin-silver mixed liquid into strips to obtain the tin-silver master alloy with a tin content of 85wt% and a silver content of 15wt%. The content of tin and silver is obtained by using a spectrometer (brand name: SPECTRO, Germany; model number: SPECTROLAB S) to test a small amount of the object to be tested which is cast into ingot form from the tin-silver mixed solution.

The preparation steps of the tin-nickel master alloy are similar to those of the tin-silver master alloy, except that the silver metal in step (1) is replaced with nickel metal (purchased from Norway) to obtain the tin-nickel master alloy with a tin content of 99 wt% and a nickel content of 1 wt%.

The preparation steps of the tin-copper master alloy are similar to those of the tin-silver master alloy, except that the silver metal in step (1) is replaced with copper metal (purchased from Japan) to obtain the tin-copper master alloy with a tin content of 95wt% and a copper content of 5wt%.

The preparation steps of the tin-germanium master alloy are similar to those of the tin-silver master alloy, except that the silver metal in step (1) is replaced with germanium metal (purchased from the United States) to obtain the tin-germanium master alloy with a tin content of 99 wt% and a germanium content of 1 wt%.

The preparation steps of the tin-bismuth master alloy are as follows: (1). Pure tin (purchased from Timah, Indonesia, with a purity of 3N) is placed in a melting furnace and heated to 400°C to completely melt the pure tin into a liquid state, thereby forming a tin melt; (2). The tin melt is stirred for 10 minutes using the stirrer, and then bismuth metal (purchased from Taiwan) is placed in the melting furnace and stirred for 30 minutes to form a tin-bismuth mixed solution; and (3). The tin-bismuth mixed solution is cast into a strip to obtain the tin-bismuth master alloy with a tin content of 80wt% and a bismuth content of 20wt%. The contents of tin and bismuth are obtained by using the spectrometer to test a small amount of the test object cast into a tablet from the tin-bismuth mixed solution.

The preparation steps of the tin-indium master alloy are similar to those of the tin-bismuth master alloy, except that the bismuth metal in step (2) is replaced with indium metal (purchased from Taiwan) to obtain the tin-indium master alloy with a tin content of 95wt% and an indium content of 5wt%.

The preparation steps of the tin-phosphorus master alloy are similar to those of the tin-bismuth master alloy, except that the bismuth metal in step (2) is replaced with phosphorus metal (purchased from India) to obtain a tin-phosphorus master alloy with a tin content of 99 wt% and a phosphorus content of 1 wt%.

〈Implementation Example 1 〉

The preparation steps of the tin-based alloy are as follows: (1). Pure tin (purchased from Timah, Indonesia, with a purity of 3N) is placed in the melting furnace and heated to 400°C to completely melt the pure tin into a liquid state, thereby forming a tin melt; (2). The tin melt is stirred for 10 minutes using the stirrer, and then the prepared master alloy (specifically the tin-silver master alloy, the tin-nickel master alloy, the tin-bismuth master alloy and the tin-phosphorus master alloy) is placed in the melting furnace and stirred for 90 minutes to form a tin-based mixed liquid; and (3). After adjusting the temperature of the melting furnace to 300°C, the tin-based mixed liquid is cast into strips to obtain the tin-based alloy with a tin content of 96.4475wt%, a silver content of 2wt%, a bismuth content of 1.5wt%, a nickel content of 0.05wt% and a phosphorus content of 0.0025wt%. The contents of tin, silver, bismuth, nickel and phosphorus are obtained by using the spectrometer to detect a small amount of the test object cast into ingots from the tin-based mixed liquid.

〈Implementation Example 2 to 16 and Comparison Examples 1 to 12 〉

The tin-based alloys of Examples 2 to 16 and Comparative Examples 1 to 12 were prepared in a manner similar to that of Example 1, except that the master alloy used was changed so that the contents of the components in the prepared tin-based alloys were changed as shown in Tables 3 and 5.

〈Evaluation Items〉

The tin-based alloys of Examples 1 to 15 and Comparative Examples 1 to 6 were evaluated as follows. For the sake of clarity, the following evaluation items are described using the tin-based alloy of Example 1 as an example. The remaining Examples and Comparative Examples are analyzed according to the same testing process. The evaluation results are shown in Table 3.

Thrust test: Evaluation of interface fracture patterns

Take the tin-based alloy of Example 1 to make a solder ball with a diameter of 220µm. Prepare a solder pad (specifically a copper substrate) that includes an opening with a diameter of 200µm. After placing the solder ball in the opening, reflow the solder ball three times to form a solder joint on the solder pad. Each reflow is to perform a heating and cooling cycle on the solder ball in a nitrogen environment. The heating and cooling cycle is to heat the solder ball from room temperature (25°C), maintain the temperature to the maximum value (245±5°C) for 40 seconds to 60 seconds, and then cool it back to room temperature at a rate of 1°C to 3°C per second.

According to the JESD22-B117B (2014 edition) specification, a push test was performed on the solder joint using a push-pull test machine (brand: Dage; model: Dage-4000), and a microscope (brand: Olympus; model: SZ51) was used to observe a section formed by the fracture of the solder joint to confirm the solder joint residue and whether the interface metal layer of the section was intact. Then, the interface fracture type of the solder joint formed by the tin-based alloy in Example 1 was evaluated according to Table 1.

Table 1 Evaluation criteria for interface fracture patterns Cross-section Residue of solder joint on cross section Is the metal layer intact? Evaluation results Symbol Ductile fracture of solder joints 100% yes Excellent ◎ Brittle fracture of solder joints 100% yes good ○ Pad fracture 0% yes Fair △ Fracture of the intermetallic layer Less than 25% no Poor ✕

Temperature Cycle Test: Evaluation of Thermal Fatigue Resistance

A tin ball with a diameter of 220 μm is made from the tin-based alloy of Example 1, and then, according to the JESD22-B111 (2003 edition) specification, an electronic component is soldered to a pad (specifically a copper substrate) using the tin ball to obtain a test product, the test product comprising the electronic component, the pad, and a solder joint disposed between the electronic component and the pad.

According to the IPC-9701 (2006 edition) specification, the test object is placed in a temperature cycle environment for a temperature cycle test, and the electrical connection signal of the test object is detected by an instrument. When the instrument cannot detect the electrical connection signal of the test object, it means that the solder joint has failed in electrical connection. The total number of temperature cycles experienced by the test object at this time is recorded, and the thermal fatigue resistance of the solder joint formed by the tin-based alloy of Example 1 is evaluated according to Table 2. Each temperature cycle includes a heating stage of 15 minutes, a high temperature dwell stage of 15 minutes, a cooling stage of 15 minutes and a low temperature dwell stage of 15 minutes, and takes room temperature (25°C) as the starting temperature, the temperature of the high temperature dwell stage is 125°C, and the temperature of the low temperature dwell stage is -40°C.

Table 2 Evaluation criteria for heat fatigue resistance Total number of temperature cycles Evaluation results Symbol More than 1500 times Excellent ◎ 1001 to 1500 good ○ 501 to 1000 times Fair △ Less than 500 times Poor ✕

Composition detection of intermetallic layer

Take the tin-based alloy of Example 1 to make a tin ball with a diameter of 220µm. Prepare a solder pad (specifically a nickel-gold substrate or a copper substrate), which includes an opening with a diameter of 200µm. After placing the tin ball in the opening, reflow the tin ball three times to form a solder joint on the solder pad, and obtain a test piece to be tested including the solder joint. Each reflow is to perform a heating and cooling cycle on the tin ball in a nitrogen environment. The heating and cooling cycle is to heat the tin ball from room temperature (25°C), maintain the temperature to the maximum value (245±5°C) for 40 seconds to 60 seconds, and then cool it back to room temperature at a rate of 1°C to 3°C per second.

The following operations were performed on the test piece in sequence: cold mounting with resin, grinding with sandpaper, and polishing with alumina polishing liquid to obtain a processed test piece. The intermetallic layer of the solder joint in the processed test piece was observed using a scanning electron microscope (SEM) (brand name: Phenom; model name: ProX), and the composition of the intermetallic layer was analyzed using an energy dispersive X-ray analysis (EDX) (brand name: Phenom; model name: ProX).

Melting point detection of tin-based alloys

A tin ball with a diameter of 220 μm was prepared from the tin-based alloy of Example 1. The tin ball was placed in a differential scanning calorimeter (DSC) (PerkinElmer; model DSC 4000), and the temperature at which the tin ball began to melt was recorded.

Table 3 Tin-based alloy content (wt%) Evaluation items Lead Copper Germanium silver Bismuth Nickel phosphorus Indium Tin Copper substrate Nickel gold substrate Tin-based alloy Interface fracture type Heat fatigue resistance Intermediate metal layer composition Melting point(℃) Embodiment 1 0 0 0 2 1.5 0.05 0.0025 0 Remaining amount ◎ ○ Ni ₃ Sn ₄ 218.30 2 0 0 0 3 1.5 0.05 0.0025 0 Remaining amount ◎ ○ Ni ₃ Sn ₄ 218.12 3 0 0 0 4 1.5 0.05 0.0025 0 Remaining amount ◎ ○ Ni ₃ Sn ₄ 218.22 4 0 0 0 5 1 0.05 0.0025 0 Remaining amount ◎ ○ Ni ₃ Sn ₄ 218.86 5 0 0 0 5 1.5 0.05 0.0025 0 Remaining amount ◎ ◎ Ni ₃ Sn ₄ 218.11 6 0 0 0 5 2 0.05 0.0025 0 Remaining amount ○ △ Ni ₃ Sn ₄ 216.61 7 0 0 0 5 1.5 0.02 0.0025 0 Remaining amount ○ △ Ni ₃ Sn ₄ 218.20 8 0 0 0 5 1.5 0.04 0.0025 0 Remaining amount ◎ △ Ni ₃ Sn ₄ 218.10 9 0 0 0 5 1.5 0.06 0.0025 0 Remaining amount ◎ ◎ Ni ₃ Sn ₄ 218.30 10 0 0 0 5 1.5 0.07 0.0025 0 Remaining amount ◎ ◎ Ni ₃ Sn ₄ 218.40 11 0 0 0 5 2 0.05 0.0025 0.25 Remaining amount ◎ ◎ Ni ₃ (SnIn) ₄ 214.11 12 0 0 0 5 2 0.05 0.0025 0.5 Remaining amount ◎ ◎ Ni ₃ (SnIn) ₄ 213.40 13 0 0 0 5 2 0.05 0.0025 0.75 Remaining amount ◎ ◎ Ni ₃ (SnIn) ₄ 212.70 14 0 0 0 5 2 0.05 0.0025 1 Remaining amount ◎ ◎ Ni ₃ (SnIn) ₄ 212.00 15 0 0 0 5 2 0.05 0.0025 1.5 Remaining amount ◎ ◎ Ni ₃ (SnIn) ₄ 210.60 Comparative example 1 0 0 0 6 1.5 0.05 0.0025 0 Remaining amount ✕ ✕ Ni ₃ Sn ₄ 218.90 2 0 0 0 5 3 0.05 0.0025 0 Remaining amount ✕ ◎ Ni ₃ Sn ₄ 214.57 3 0 0 0 5 1.5 0 0.0025 0 Remaining amount ✕ △ Ni ₃ Sn ₄ 217.69 4 0 0 0 5 1.5 0.08 0.0025 0 Remaining amount ✕ ○ Ni ₃ Sn ₄ 217.90 5 0 0.5 0 3 0 0.05 0.0025 0 Remaining amount ◎ △ (CuNi) ₆ Sn ₅ (NiCu) ₃ Sn ₄ 217.60 6 0 0.5 0 4 2 0.05 0.0025 0 Remaining amount ✕ ◎ (CuNi) ₆ Sn ₅ (NiCu) ₃ Sn ₄ 212.40

Referring to Table 3, the solder joints formed by the tin-based alloys of Examples 1 to 10 do have both good or even excellent interface fracture morphology and acceptable thermal fatigue resistance. In particular, the tin-based alloys of Examples 11 to 15 further contain 0.2 wt % to 1.5 wt % of indium, so that the solder joints formed have both excellent interface fracture morphology and thermal fatigue resistance. In addition, the solder joints formed by the tin-based alloys of Examples 1 to 10 have an intermetallic layer formed by Ni ₃ Sn _4. Since the formed solder joints have only one intermetallic layer, the solder joints are less likely to suffer from electrical connection failure. The solder joints formed by the tin-based alloys of Examples 11 to 15 have an intermetallic layer formed by Ni ₃ (SnIn) ₄ with a fine structure, a thin thickness, and a non-brittleness. Therefore, the solder joints also have the advantage of being non-fracture prone, and thus the solder joints are less likely to suffer from electrical connection failure. Furthermore, the melting point of the tin-based alloys of Examples 1 to 15 is between 210°C and 222°C, which is close to the melting point of the tin-silver-copper alloy commonly used in the industry (about 217°C). Therefore, they are applicable to the reflow curve of the tin-silver-copper alloy, so there is no need to re-adjust the reflow curve during use, and no production cost derived from adjusting the reflow curve will be generated.

Compared to the tin-based alloy of comparative example 1, which has poor interface fracture strength and thermal fatigue resistance of the formed solder joint due to the silver content being greater than 5wt%, the tin-based alloys of embodiments 1 to 15 have better interface fracture strength and thermal fatigue resistance of the formed solder joint due to the silver content being between 2wt% and 5wt%.

Compared to the tin-based alloy of comparative example 2, which has poor interfacial fracture strength of the formed solder joint due to the bismuth content being greater than 2wt%, the tin-based alloys of embodiments 1 to 15 have better interfacial fracture morphology of the formed solder joint due to the bismuth content being between 0.1wt% and 2wt%.

Compared with the tin-based alloy of Comparative Example 3, which does not contain nickel and thus has poor interfacial fracture strength of the formed solder joint, and compared with the tin-based alloy of Comparative Example 4, which has poor interfacial fracture strength of the formed solder joint due to the nickel content being greater than 0.07wt%, the tin-based alloys of Examples 1 to 15 have a nickel content between 0.02wt% and 0.07wt% and thus have formed solder joints with better interfacial fracture morphology.

Compared to the tin-based alloys of Comparative Examples 5 and 6, which contain copper and thus have two intermetallic layers of (CuNi) ₆ Sn ₅ and (NiCu) ₃ Sn ₄ in the formed solder joints, the tin-based alloys of Examples 1 to 15 do not contain copper and thus have only one intermetallic layer in the formed solder joints. Furthermore, compared to the tin-based alloy of Comparative Example 6, which has poor interfacial fracture strength in the formed solder joints due to the copper content, the tin-based alloys of Examples 1 to 15 do not contain copper and thus have better interfacial fracture strength in the formed solder joints.

The tin-based alloys of Examples 3 and 16 and Comparative Examples 7 to 12 were evaluated as follows. For the sake of clarity, the following evaluation items are described using the tin-based alloy of Example 3 as an example, and the remaining Examples and Comparative Examples are analyzed according to the same testing process. The evaluation results are shown in Table 5.

Evaluation of antioxidant activity

Three groups of 30g tin balls were prepared from the tin-based alloy of Example 3. Then, one group of tin balls was placed in an oven at a temperature of 240±5°C for a 10-minute high-temperature baking test. Another group of tin balls was placed in a constant temperature and humidity machine (brand: GIANT FORCE; model: GTH-120-00-SP-AR) at a temperature of 85°C and a humidity of 85% for a 24-hour high-temperature and high-humidity test. The remaining group of tin balls was used as a control group. The oxidation resistance of the tin balls formed from the tin-based alloy of Example 3 was evaluated according to the Pantone international color card and Table 4.

Compared with the solder ball surface of the control group, if the evaluation result of the solder ball surface in the high temperature baking test is good and the evaluation result of the solder ball surface in the high temperature and high humidity test is good, it means that the oxidation resistance of the solder ball formed by the tin-based alloy is "good"; if the evaluation result of the solder ball surface in at least one of the high temperature baking test and the high temperature and high humidity test is fair, it means that the oxidation resistance of the solder ball is "fair"; if the evaluation result of the solder ball surface in at least one of the high temperature baking test and the high temperature and high humidity test is poor, it means that the oxidation resistance of the solder ball is "poor".

Table 4 Evaluation criteria for antioxidant activity High temperature baking test High temperature and high humidity test The surface of the tin ball Evaluation results Symbol The surface of the tin ball Evaluation results Symbol No oxidation/no discoloration good ○ No oxidation/no discoloration good ○ Brightness reduction Fair △ Brightness reduction Fair △ Darker color Poor ✕ Darker color Poor ✕

Table 5 Tin-based alloy content (wt%) Evaluation items Lead Copper Germanium silver Bismuth Nickel phosphorus Tin High temperature baking test High temperature and high humidity test Embodiment 3 0 0 0 4 1.5 0.05 0.0025 Remaining amount ○ ○ 16 0 0 0 4 1.5 0.05 0.0055 Remaining amount ○ ○ Comparative example 7 0 0 0 4 1.5 0.05 0 Remaining amount ✕ ✕ 8 0 0 0 4 1.5 0.05 0.0015 Remaining amount ✕ ✕ 9 0 0 0 4 1.5 0.05 0.0070 Remaining amount △ △ 10 0 0 0.005 4 1.5 0.05 0 Remaining amount ✕ ✕ 11 0 0 0.01 4 1.5 0.05 0 Remaining amount △ △ 12 0 0 0.02 4 1.5 0.05 0 Remaining amount ○ ○

Referring to Table 5, the evaluation results of the tin balls formed by the tin-based alloys of Examples 3 and 16 in the high temperature baking test and the high temperature and high humidity test are both good, indicating that the tin balls formed by the tin-based alloys of Examples 3 and 16 have good oxidation resistance.

Compared with the tin-based alloys of Comparative Examples 7 and 8, which have a phosphorus content of less than 0.002wt%, the oxidation resistance of the formed tin balls is poor, and compared with the tin-based alloy of Comparative Example 9, which has a phosphorus content of more than 0.006wt%, the oxidation resistance of the formed tin balls is fair. The tin-based alloys of Examples 3 and 16 have a phosphorus content between 0.002wt% and 0.006wt%, so the formed tin balls have better oxidation resistance.

The tin-based alloys of Comparative Examples 10 to 12 are lead-free and copper-free tin alloys disclosed in Taiwan Patent Publication No. 789165. Compared with the tin-based alloys of Comparative Examples 10 to 11, which do not contain phosphorus but contain germanium in a content range of 0.005wt% to 0.01wt%, so that the oxidation resistance of the formed tin balls is fair or even poor, the tin-based alloys of Examples 3 and 16 do not contain germanium but contain phosphorus in a content range of 0.002wt% to 0.006wt%, so that the formed tin balls have better oxidation resistance.

The tin-based alloy of Comparative Example 12 does not contain phosphorus but contains germanium in an amount of 0.02 wt %, which enables the formed tin ball to have good oxidation resistance. The tin-based alloys of Examples 3 and 16 do not contain germanium but contain phosphorus in an amount ranging from 0.002 wt % to 0.006 wt %, which enables the formed tin ball to have good oxidation resistance. It can be seen that germanium or phosphorus at a specific content can give the tin ball good oxidation resistance, and because the required phosphorus content is much lower than the required germanium content, and the procurement cost of phosphorus is much lower than the procurement cost of germanium metal, the tin-based alloy and the tin ball of the present invention also have the advantage of low raw material cost.

In summary, the tin-based alloy of the present invention does not contain lead, copper and germanium, and through the cooperation of 2wt% to 5wt% of silver, 0.1wt% to 2wt% of bismuth, 0.02wt% to 0.07wt% of nickel, 0.002wt% to 0.006wt% of phosphorus and the balance of tin, the tin ball of the present invention formed by the tin-based alloy has good oxidation resistance; a solder joint formed by the tin ball after multiple reflows has only one intermetallic layer, so that the solder joint is not prone to electrical connection failure; the solder joint has good or even excellent welding strength and interface fracture morphology, and maintains a certain level of thermal fatigue resistance, so that the solder joint has high reliability, so the purpose of the present invention can be achieved. At the same time, the tin-based alloy is suitable for the reflow curve of the tin-silver-copper alloy commonly used in the industry without incurring production costs derived from adjusting the reflow curve.

It is worth mentioning that the tin-based alloy of the present invention further contains 0.2wt% to 1.5wt% of indium, so that the formed solder joint has only one layer of fine structure, thin thickness and non-brittle interface metal layer, so that the solder joint also has the advantage of being non-fracture-resistant, and then the solder joint is less likely to have electrical connection failure, and the solder joint has both excellent interface fracture morphology and thermal fatigue resistance.

However, the above is only an example of the present invention and should not be used to limit the scope of implementation of the present invention. All simple equivalent changes and modifications made according to the scope of the patent application of the present invention and the content of the patent specification are still within the scope of the present invention.

Other features and effects of the present invention will be clearly presented in the embodiments with reference to the drawings, wherein: FIG. 1 is a SEM image of a tin ball formed by a tin-based alloy of Example 6 after multiple reflows on a nickel-gold substrate; FIG. 2 is a SEM image of a tin ball formed by a tin-based alloy of Example 6 after multiple reflows on a copper substrate; FIG. 3 is a SEM image of a tin ball formed by a tin-based alloy of Example 11 after multiple reflows on the nickel-gold substrate; and FIG. 4 is a SEM image of a tin ball formed by a tin-based alloy of Example 11 after multiple reflows on the copper substrate.

Claims (4)

A tin-based alloy comprises, based on 100wt% of the total amount of the tin-based alloy, 2wt% to 5wt% of silver, 0.1wt% to 2wt% of bismuth, 0.02wt% to 0.07wt% of nickel, 0.002wt% to 0.006wt% of phosphorus, and the balance of tin; and the tin-based alloy does not contain lead, copper, and germanium. The tin-based alloy as described in claim 1, wherein the tin-based alloy further contains 0.2 wt % to 1.5 wt % of indium. The tin-based alloy as described in claim 1, wherein the melting point of the tin-based alloy ranges from 210°C to 222°C. A tin ball is formed from the tin-based alloy as described in any one of claims 1 to 3.