Solder alloys, solder powders, solder pastes, solder balls, solder preforms, and solder joints

Abstract

The present invention provides a solder alloy, solder powder, solder paste, solder ball, solder preform, and solder joint that have excellent heat cycle resistance, suppress the occurrence of lift-off, and further have excellent wettability and high tensile strength. The solder alloy has an alloy composition, in mass %, of 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.5% Sb, 0.050 to 0.200% Ni, and the remainder Sn.

Description

[Technical Field]

[0001] The present invention relates to a solder alloy, a solder powder, a solder paste, a solder ball, a solder preform, and a solder joint. [Background technology]

[0002] In recent years, automobiles have become more functional and high-performance due to the electrification of mechanical parts. To achieve this, electronics packaging technology that supports miniaturization and high density is required. To support this packaging technology, high reliability is required for the solder joints that connect printed circuit boards and electronic components.

[0003] For example, printed circuit boards mounted near the engine must be used for long periods of time in environments with significant temperature differences. Hybrid vehicles are equipped with inverters that handle large amounts of power, and require solder joints that can withstand the inverter's operating temperatures. As such, automotive electronics packaging technology requires high reliability from the solder alloys that make up the solder joints, and various studies are being conducted.

[0004] Patent Document 1 discloses a Sn-Ag-Cu-Bi-Sb-Ni solder alloy that can withstand a heat cycle test of 3000 cycles at temperatures of -40°C and 125°C, each of which is maintained for 30 minutes.

[0005] Incidentally, mounting boards, which are printed circuit boards with electronic components mounted on them, are used in home appliances such as washing machines, refrigerators, and air conditioners, as well as in electronic devices such as televisions, videos, radios, computers, copiers, and communication devices. In addition to single-layer mounting boards, mounting boards made by stacking multiple boards are also used to achieve a wide range of functions.

[0006] For electrical continuity between boards or for mounting electronic components on boards, methods such as surface mounting and inserting terminals into through-holes in the board are available. Examples of such mounting processes on printed circuit boards include flow soldering, reflow soldering, and manual soldering.

[0007] Among these, the mounting method for electronic components of a certain size is to insert terminals into through holes and form fillets, taking into account factors such as connection strength. Reflow soldering is used as the mounting process. Furthermore, even these consumer products are required to be able to withstand a certain degree of heat cycle testing, given the increasing temperature fluctuations and the large currents that flow due to higher performance in recent years.

[0008] Patent Document 2 discusses a solder alloy that can withstand a 2000-cycle heat cycle test, with holding times at -30°C and 120°C for 20 minutes each, and that suppresses lift-off, which creates a gap between the Cu land and fillet of a through-hole. This document discloses a Sn-Ag-Cu-Sb-Ni-In solder alloy that contains In instead of Bi.

[0009] Patent Document 3 also studies a solder alloy that can withstand a heat cycle test of 3,000 cycles, with holding times at -40°C and 150°C for 30 minutes each, and that suppresses the occurrence of lift-off. This document discloses a Sn-Ag-Cu-Bi-Sb-Ni solder alloy that contains a small amount of Bi. [Prior art documents] [Patent documents]

[0010] [Patent Document 1] Patent No. 5811304 [Patent Document 2] Patent No. 6349615 [Patent Document 3] Patent No. 6719443 Summary of the Invention [Problem to be solved by the invention]

[0011] The heat cycle resistance characteristics of the Sn-Ag-Cu-Bi-Sb-Ni solder alloy described in Patent Document 1 are investigated. However, the solder alloys described in these documents are not considered for use in through-holes, and are not solder alloys for suppressing lift-off.

[0012] Furthermore, the solder alloy described in Patent Document 2 is said to have excellent heat cycle resistance and be able to suppress the occurrence of lift-off by containing In instead of Bi. However, as described in Patent Document 1, it is known that the inclusion of Bi together with Sb improves heat cycle resistance. For this reason, the solder alloy described in Patent Document 2 must not contain Bi, and therefore is thought to have poor heat cycle resistance even if the occurrence of lift-off is suppressed.

[0013] Patent Document 3 describes a Sn-Ag-Cu-Bi-Sb-Ni solder alloy that contains 0.1 to 0.7 mass% Bi, which is said to improve strength while suppressing lift-off. The document explains that lift-off occurs because, in solder alloys with a high Bi content, Bi tends to accumulate at the interface with the Cu land inside the fillet, and the high concentration of Bi accumulated at the interface inhibits solidification, causing lift-off. For this reason, Patent Document 3 requires that the Bi content be reduced as much as possible to suppress lift-off, and therefore the Bi content is kept small as described above.

[0014] However, the solder alloy described in Patent Document 3 inevitably suffers from a deterioration in heat cycle resistance when the Bi content is suppressed, similar to the solder alloy described in Patent Document 2. Furthermore, a low Bi content reduces the strength of the solder alloy, which may also cause a deterioration in heat cycle resistance.

[0015] Furthermore, wettability has not been considered for the solder alloy described in Patent Document 3. When forming a fillet, reflow soldering is performed after inserting the terminal into the through-hole, so excellent wettability is an extremely important property. However, since the solder alloy described in Patent Document 3 that does not contain Bi has been shown to have excellent wettability and spreadability, it is better not to include Bi in order to improve wettability. On the other hand, if Bi is not included, heat cycle resistance will be poor, as mentioned above.

[0016] Thus, although the Sn-Ag-Cu-Bi-Sb-Ni solder alloys disclosed in Patent Document 1 and Patent Document 3 contain the same constituent elements, none of them exhibited excellent heat cycle resistance, suppressed lift-off, high strength, and excellent wettability. This is thought to be because previous studies had only focused on specific effects for solder alloys used in mixed mounting on boards that combine surface mounting and through-hole insertion mounting. Therefore, there is an urgent need to provide a solder alloy that possesses all of these properties for consumer printed circuit boards in addition to automotive boards.

[0017] An object of the present invention is to provide a solder alloy, a solder powder, a solder paste, a solder ball, a solder preform, and a solder joint that have excellent heat cycle resistance, suppress the occurrence of lift-off, and further have excellent wettability and high tensile strength. [Means for solving the problem]

[0018] The present inventors randomly selected alloy compositions disclosed in Patent Documents 1 to 3 and re-examined the issues associated with each. Example 17, a Sn-3.4Ag-0.7Cu-2.5Bi-2.0Sb-0.040Ni solder alloy, was selected from Patent Document 1, which examines heat cycle resistance (figures represent mass %, the same applies below). It was discovered that lift-off occurs with this solder alloy.

[0019] Furthermore, Examples 16, 23, 30, and 37-38, which are Sn-3.4Ag-0.7Cu-1.5-2.0Bi-2.0-5.0Sb-0.040Ni solder alloys, were extracted from Patent Document 1. It was found that lift-off occurred in all of these solder alloys. Furthermore, Reference Example 9, which is a Sn-3.4Ag-0.7Cu-1.5Bi-1.0Sb-0.040Ni solder alloy, was extracted from Patent Document 1. It was found that this solder alloy suppressed the occurrence of lift-off, but had poor heat cycle resistance and tensile strength.

[0020] Because lift-off has not been investigated for these solder alloys, we investigated the relationship between these elements and lift-off. Lift-off is thought to occur when the temperature is lowered after melting during reflow soldering. Even when the Bi content is significantly lower than the eutectic composition with Sn, it forms a SnBi eutectic phase, which has a low melting point. As a result, the temperature difference ΔT between the liquidus and solidus temperatures increases as the temperature is lowered, and the molten solder does not completely solidify when the temperature is lowered. Therefore, it is presumed that stress is generated due to differences in the thermal expansion coefficients of the substrate, Cu land, and molten solder, causing the unsolidified solder to peel off from the electrode, resulting in lift-off.

[0021] According to this speculation, lift-off can be suppressed by reducing the Bi content to less than 2.5% by mass. However, lift-off occurred even in solder alloys with a Bi content of 1.5 to 2.0% by mass. This is thought to be due to the coarse Sn crystal grains and the formation of an SnBi eutectic phase, which causes the solder to not solidify completely and is then pulled off from the electrode.

[0022] More specifically, this is thought to be due to the following. Figure 1 is a schematic diagram of the alloy structure of a solder alloy, with Figure 1(a) showing an alloy structure with coarse Sn crystal grains and Figure 1(b) showing an alloy structure with fine Sn crystal grains. As shown in Figure 1(a), when the content of Ni, which functions as a crystal nucleus, is low, the Sn crystal grains become coarse and the specific surface area of ​​the grain boundaries becomes small. As a result, the concentration of Bi in the specific surface area exceeds the solid solubility limit, and the alloy composition at the grain boundaries approaches an SnBi eutectic composition. The eutectic composition is formed partially in the solder alloy as a composition that cannot be completely solidified by the eutectic reaction, which solidifies at the lowest temperature. As a result, the solder alloy cannot withstand the stress of being pulled away from the electrode due to the difference in linear expansion coefficients during solidification, resulting in lift-off.

[0023] On the other hand, as shown in Figure 1(b), when the Ni content is high, the Sn crystal grains become finer and the specific surface area of ​​the grain boundaries increases. As a result, the concentration of Bi that exceeds the solid solubility limit in the specific surface area decreases, the alloy composition of the grain boundaries does not approach the Sn-Bi eutectic composition, and the eutectic reaction during solidification is suppressed. This makes it possible to suppress the occurrence of lift-off. Thus, it is thought that lift-off requires control of not only the Bi content but also the Ni content.

[0024] Furthermore, from the viewpoint of controlling the solid solubility limit of Bi, the inventors also focused on Sb, which, like Bi, is solid-soluble in Sn. Figure 2 shows the relationship between the Sb content and the solid solubility limit of Bi and the alloy structure, with Figure 2(a) being a graph showing the relationship between the Sb content and the solid solubility limit of Bi, and Figure 2(b) being a schematic diagram showing the relationship between the Sb content and the alloy structure. As shown in Figure 2(a), when the Sb content is reduced, the precipitation of SnSb is significantly suppressed and the solid solubility limit of Bi in Sn increases.

[0025] As shown in Figure 2(b), when the Sb content is high, a SnBi eutectic phase is formed on the electrode, whereas when the Sb content is low, the solubility limit of Bi in Sn increases, suppressing the formation of the SnBi eutectic phase. From the above, it is thought that the Sb content also needs to be adjusted to suppress the occurrence of lift-off.

[0026] As described above, it is clear that the occurrence of lift-off can be suppressed by reducing the contents of Bi and Sb and increasing the content of Ni. However, it is believed that reducing the contents of Bi and Sb will deteriorate the heat cycle resistance and tensile strength. Furthermore, it is believed that a high Ni content will deteriorate the wettability. Therefore, it is believed that the effects of the present invention can be simultaneously achieved by adjusting the contents of Bi, Sb, and Ni to such an extent that lift-off is suppressed.

[0027] Next, Example 1, a Sn-3.4Ag-0.7Cu-3.0Sb-0.07Ni-3.0In-0.01Ge solder alloy, was extracted from Patent Document 2, which also examines lift-off. Similarly, Example 3, a Sn-3.0Ag-0.7Cu-0.7Bi-3.0Sb-0.03Ni solder alloy, and Reference Example 15, a Sn-3.0Ag-0.7Cu-0.7Bi-3.5Sb-0.15Ni solder alloy, were extracted from Patent Document 3, which also examines lift-off. It was discovered that these solder alloys have poor wettability. Furthermore, both Example 1 of Patent Document 2 and Example 3 of Patent Document 3 were poor in wettability as well as heat cycle resistance and tensile strength.

[0028] Among these, the solder alloy of Example 1 in Patent Document 2 does not contain Bi, and therefore can suppress the occurrence of lift-off. However, the absence of Bi leads to insufficient solid solution strengthening, and it is presumed that this leads to poor heat cycle resistance and tensile strength. Furthermore, since Bi is an element that reduces surface tension, solder alloys that do not contain Bi are also presumed to have poor wettability. The solder alloy of Example 3 in Patent Document 3 not only has poor wettability due to its low Bi content, but also has a low Ni content, which prevents Sn crystal grains from becoming finer, presumed to result in poor heat cycle resistance and tensile strength. Reference Example 15 in Patent Document 3 is presumed to have poor wettability due to its low Bi content.

[0029] Thus, in solder alloys that suppress lift-off, the heat cycle resistance, wettability, and tensile strength are poor, so it seems that increasing the Bi content would be sufficient, but as mentioned above, increasing the Bi content leads to lift-off.

[0030] Therefore, the present inventors conducted detailed studies on all of the constituent elements to find a solder alloy that simultaneously exhibits all of the above-mentioned effects. As a result, they discovered that in a Sn-Ag-Cu-Bi-Sb-Ni solder alloy, these effects can be exhibited at a practically acceptable level only when the contents of each constituent element are within a predetermined range. That is, they discovered that by adjusting the Bi content, and then adjusting the Ag and Cu contents in addition to Sb and Ni, it is possible to achieve a solder alloy that suppresses the occurrence of lift-off, has excellent heat cycle resistance and wettability, and has high tensile strength, and thus completed the present invention. The present invention, which was made based on these findings, is as follows.

[0031] (0) A solder alloy characterized by comprising, in mass %, 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.5% Sb, 0.050 to 0.200% Ni, and the remainder being Sn. (1) A solder alloy characterized by having an alloy composition, in mass%, of 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.5% Sb, 0.050 to 0.200% Ni, and the remainder being Sn.

[0032] (2) A solder alloy characterized by having an alloy composition, in mass%, of 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.0% Sb, 0.060 to 0.190% Ni, 0.3% or less Co, and the remainder being Sn. A solder alloy characterized by comprising, in mass %, 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.0% Sb, 0.060 to 0.190% Ni, 0.3% or less Co, and the remainder being Sn.

[0033] (3) The solder alloy according to any one of (0) to (2) above, further containing, by mass %, 0.11% or less in total of at least one of Ge, Ga, As, Fe, Pd, Mn, In, Zn, Zr, and Mg.

[0034] (4) The solder alloy according to any one of (0) to (3) above, which satisfies the following formulas (1) and (2): 0.216≦(Ag×Cu) / (Bi×Sb)≦0.465 (1) 34.2≦Bi×Sb / Ni≦85.6 (2) In the above formulas (1) and (2), Ag, Cu, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

[0035] (5) The solder alloy according to (4) above, which satisfies the following formula (3): 1.11≦Ag×Bi×Sb×Ni≦5.84 (3) In the above formula (3), Ag, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

[0036] (6, 7) A solder powder made of the solder alloy according to any one of (0) to (5) above.

[0037] (8, 9) A solder paste containing the solder powder according to (6) or (7) above.

[0038] (10, 11) A solder ball made of the solder alloy according to any one of (0) to (5) above.

[0039] (12, 13) A solder preform made of the solder alloy according to any one of (0) to (5) above.

[0040] (14, 15) A soldered joint comprising the solder alloy according to any one of (0) to (5) above. [Brief explanation of the drawings]

[0041] [Figure 1] Figure 1 is a schematic diagram of the alloy structure of a solder alloy, where Figure 1(a) shows an alloy structure in which coarse Sn crystal grains are precipitated, and Figure 1(b) shows an alloy structure in which fine Sn crystal grains are precipitated. [Figure 2] FIG. 2 shows the relationship between the Sb content and the solid solubility limit of Bi and the alloy structure. FIG. 2(a) is a graph showing the relationship between the Sb content and the solid solubility limit of Bi, and FIG. 2(b) is a schematic diagram showing the relationship between the Sb content and the alloy structure. [Figure 3] Figure 3 shows cross-sectional SEM photographs of the fillet, with Figure 3(a) being an SEM photograph enlarged 30 times and Figure 3(b) being an SEM photograph enlarged 300 times. [Figure 4] FIG. 4 shows the cracks 1 and 2 actually observed in a cross-sectional SEM photograph of the fillet, as well as the predicted crack propagation paths from these cracks. [Figure 5] Figure 5 is an SEM photograph of the cross section of a fillet magnified 300 times, where Figure 5(a) is an SEM photograph of a fillet formed using Comparative Example 17, and Figure 5(b) is an SEM photograph of a fillet formed using Example 3. [Figure 6] 6A and 6B are graphs showing the relationship between the temperature and the solid fraction obtained from the DSC curves during cooling, where FIG. 6A is the graph for Example 3 and FIG. 6B is the graph for Comparative Example 13. DETAILED DESCRIPTION OF THE INVENTION

[0042] The present invention will be described in more detail below. In this specification, "%" relating to the solder alloy composition means "% by mass" unless otherwise specified.

[0043] 1. Solder alloy (1) Ag: 3.0-4.0% Ag improves tensile strength and heat cycle resistance through precipitation strengthening of Ag3Sn. Furthermore, Ag also improves wettability. If the Ag content is less than 3.0%, the tensile strength, heat cycle resistance, and wettability deteriorate. The lower limit of the Ag content is 3.0% or more, preferably 3.1% or more, more preferably 3.2% or more, even more preferably 3.3% or more, and particularly preferably 3.4% or more.

[0044] On the other hand, if the Ag content exceeds 4.0%, coarse Ag3Sn compounds precipitate, deteriorating tensile strength and heat cycle resistance. The upper limit of the Ag content is 4.0% or less, preferably 3.9% or less, more preferably 3.8% or less, even more preferably 3.7% or less, particularly preferably 3.6% or less, and most preferably 3.5% or less. The preferred range of the Ag content is 3.3 to 3.5%.

[0045] (2) Cu: 0.60-0.80% Cu improves tensile strength and heat cycle resistance by forming a compound with Sn. If the Cu content is less than 0.60%, the tensile strength and heat cycle resistance deteriorate. The lower limit of the Cu content is 0.60% or more, preferably 0.63% or more, more preferably 0.65% or more, even more preferably 0.68% or more, and particularly preferably 0.70% or more.

[0046] On the other hand, if the Cu content exceeds 0.80%, coarse compounds with Sn are formed, deteriorating the tensile strength and heat cycle resistance. The upper limit of the Cu content is 0.80% or less, preferably 0.78% or less, more preferably 0.75% or less, and even more preferably 0.73% or less. The preferred range of the Cu content is 0.65 to 0.75%.

[0047] (3) Bi: 0.8-2.3% Bi improves tensile strength and heat cycle resistance by solid solution strengthening Sn. Furthermore, if the Bi content is appropriate, lift-off can be suppressed. Furthermore, Bi can also improve wettability. If the Bi content is less than 0.8%, wettability decreases. Furthermore, in solder alloys that do not contain Bi, tensile strength and heat cycle resistance deteriorate in addition to wettability. The lower limit of the Bi content is 0.8% or more, preferably 1.1% or more, more preferably 1.3% or more, even more preferably 1.6% or more, particularly preferably 1.8% or more, and most preferably 1.9% or more.

[0048] On the other hand, if the Bi content exceeds 2.3%, the Bi concentration in the specific surface area exceeds the solid solubility limit, and the alloy composition at the grain boundaries approaches a Sn-Bi eutectic composition. The eutectic composition is formed partially in the solder alloy as a composition that cannot be completely solidified by the eutectic reaction, which solidifies at the lowest temperature. As a result, the solder alloy cannot withstand the stress of peeling from the electrode due to the difference in linear expansion coefficients during solidification, causing lift-off. The upper limit of the Bi content is 2.3% or less, preferably 2.2% or less, more preferably 2.1% or less, and even more preferably 2.0% or less. The preferred range of the Bi content is 1.8 to 2.2%.

[0049] (4) Sb: 1.2 to 5.5% Sb improves tensile strength by solid-solution strengthening Sn. It also improves heat cycle resistance. Furthermore, if the Sb content is appropriate, the amount of precipitated SnSb compounds does not affect lift-off, suppressing the occurrence of lift-off. In addition, wettability is improved.

[0050] If the Sb content is less than 1.2%, the heat cycle resistance will decrease. If the Sb content is even lower, Sb will not be sufficiently dissolved in Sn, resulting in insufficient solid solution strengthening and a decrease in tensile strength. The lower limit of the Sb content is 1.2% or more, more preferably 1.5% or more, even more preferably 2.1% or more, particularly preferably 2.5% or more, and most preferably 2.8% or more.

[0051] On the other hand, if the Sb content exceeds 5.5%, wettability deteriorates. Furthermore, as the Sb content increases, Sn is consumed to form compounds with Sb, reducing the Sn concentration in the liquid phase that constitutes the molten solder during solidification. This reduces the Bi concentration that can dissolve in Sn, increasing the Bi concentration exceeding the solubility limit in the specific surface area, and the alloy composition at the grain boundaries approaches a Sn-Bi eutectic composition. The eutectic composition is partially formed in the solder alloy as a composition that cannot be completely solidified by the eutectic reaction that solidifies at the lowest temperature. Therefore, the Sb content cannot withstand the stress of peeling from the electrode due to the difference in linear expansion coefficients during solidification, resulting in lift-off. The upper limit of the Sb content is 5.5% or less, more preferably 4.9% or less, even more preferably 4.4% or less, even more preferably 3.9% or less, particularly preferably 3.4% or less, and most preferably 3.2% or less. The preferred range of the Sb content is 2.8 to 3.2%.

[0052] (5) Ni: 0.050~0.200% Ni functions as a crystal nucleus when the molten solder solidifies, making the Sn crystal grains finer and increasing the specific surface area of ​​the grain boundaries. As a result, the concentration of Bi that exceeds the solubility limit in the specific surface area decreases, preventing the alloy composition of the grain boundaries from approaching the Sn-Bi eutectic composition and suppressing the eutectic reaction during solidification. This prevents lift-off from occurring. In addition, SnNi compounds that react with Sn are dispersed and precipitated in the solder alloy, which refines the structure of the solder alloy and improves tensile strength.

[0053] If the Ni content is less than 0.050%, the Sn crystal grains will not be refined, causing lift-off. Also, the tensile strength may be reduced. The lower limit of the Ni content is 0.050% or more, preferably 0.060% or more, more preferably 0.090% or more, even more preferably 0.125% or more, and particularly preferably 0.150% or more.

[0054] On the other hand, if the Ni content exceeds 0.200%, wettability deteriorates due to the precipitation of a large amount of SnNi compounds. The upper limit of the Ni content is 0.200% or less, preferably 0.190% or less, more preferably 0.175% or less, and even more preferably 0.160% or less. The preferred range of the Ni content is 0.125 to 0.175%.

[0055] (6) Co: 0.3% or less Co, like Ni, is an element that functions as a crystal nucleus during solidification of molten solder and can refine the crystal grains of Sn. The upper limit of Co is preferably 0.3% or less, more preferably 0.1% or less, even more preferably 0.05% or less, and particularly preferably 0.011% or less.

[0056] The lower limit of the Co content is not particularly limited, but in order to fully exert the effect of containing Co, the lower limit of the Co content should be 0.001% or more. The preferred range of the Co content is 0.009 to 0.011%.

[0057] In this way, in the present invention, Co compensates for Ni, so when the solder alloy according to the present invention contains Co, the Ni content is 0.060 to 0.190%. When Co is contained, the preferred range of the Ni content is the same as described above.

[0058] Furthermore, when the solder alloy according to the present invention contains Co in addition to Sb, a SnCo compound precipitates in addition to SnSb. Therefore, from the viewpoint of wettability, the Sb content is 1.2 to 5.0%. The preferred range of the Sb content when Co is contained is the same as above. Furthermore, the same applies to Ag, Cu, and Bi when Co is contained.

[0059] (7) At least one of Ge, Ga, As, Fe, Pd, Mn, In, Zn, Zr, and Mg, in total, not more than 0.11% In addition to the essential elements, the solder alloy according to the present invention may contain at least one of Ge, Ga, As, Fe, Pd, Mn, In, Zn, Zr, and Mg in a total amount of 0.11% or less, to the extent that the effects of the present invention are not impaired. There is no particular lower limit, but it is sufficient as long as it is 0.0001% or more.

[0060] (8) Remainder: Sn The balance of the solder alloy according to the present invention is Sn, and may contain inevitable impurities in addition to the above-mentioned elements. The balance of the solder alloy according to the present invention may consist of Sn and inevitable impurities. Even if inevitable impurities are contained, the above-mentioned effects are not affected.

[0061] (9) Equations (1) and (2) 0.216≦(Ag×Cu) / (Bi×Sb)≦0.465 (1) 34.2≦Bi×Sb / Ni≦85.6 (2) In the above formulas (1) and (2), Ag, Cu, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

[0062] The solder alloy according to the present invention preferably satisfies the formulas (1) and (2). Regarding formula (1), as described above, when Ag and Cu are contained in appropriate amounts, they can exhibit high tensile strength and excellent heat cycle resistance. Furthermore, when Bi and Sb are contained below the lower limits described above, the tensile strength and heat cycle resistance deteriorate. Therefore, when Ag and Cu are contained in a balanced manner, and when Bi and Sb are also contained in a balanced manner, the tensile strength and heat cycle resistance are further improved.

[0063] Regarding formula (2), as mentioned above, Bi, Sb, and Ni are all elements that contribute to improving wettability. Furthermore, if the contents of Bi and Sb do not exceed the upper limits, the occurrence of lift-off is suppressed. Furthermore, if the contents of Bi and Sb are not below the lower limits, excellent heat cycle resistance is exhibited and high tensile strength can also be exhibited. On the other hand, if the content of Ni is not below the lower limits, the occurrence of lift-off is suppressed. Therefore, if Bi, Sb, and Ni satisfy formula (2), the occurrence of lift-off is suppressed and wettability is improved, and in addition, tensile strength is further improved and a higher level of heat cycle resistance is exhibited.

[0064] (10) Equation (3) 1.11≦Ag×Bi×Sb×Ni≦5.84 (3) In the above formula (3), Ag, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

[0065] The solder alloy according to the present invention may satisfy formula (3) in addition to formulas (1) and (2). With regard to formula (3), when the essential elements of the solder alloy according to the present invention are contained in an appropriate amount overall, the effects of the present invention are more easily exhibited.

[0066] The values ​​shown in Tables 1 and 2, which are the measured values ​​of the alloy composition, were used in the calculations of formulas (1) to (3). The values ​​calculated from formulas (1) to (3) are calculated to three decimal places for formula (1), to one decimal place for formula (2), and to two decimal places for formula (3). This calculation rule is used in this application and is intended to be used in the same way for calculations regarding additional solder alloys described in other documents, etc., since all solder alloys must be treated in the same way.

[0067] The lower limit of formula (1) is preferably 0.216 or more, more preferably 0.238 or more, even more preferably 0.243 or more, still more preferably 0.305 or more, particularly preferably 0.340 or more, and most preferably 0.345 or more, 0.350 or more, 0.361 or more, 0.368 or more, 0.372 or more, or 0.385 or more. The upper limit of formula (1) is preferably 0.465 or less, more preferably 0.453 or less, even more preferably 0.441 or less, still more preferably 0.425 or less, particularly preferably 0.408 or less, and most preferably 0.397 or less. A more preferable range of the formula (1) is 0.361 to 0.441. The above upper and lower limits can each define a more preferable range of the formula (1).

[0068] The lower limit of formula (2) is preferably 34.2 or more, more preferably 36.0 or more, even more preferably 37.3 or more, and particularly preferably 40.0 or more. The upper limit of formula (2) is preferably 85.6 or less, more preferably 75.0 or less, even more preferably 73.3 or less, still more preferably 66.7 or less, particularly preferably 65.3 or less, and most preferably 52.0 or less, 48.0 or less, 46.0 or less, 45.3 or less, 44.0 or less, or 42.7 or less. A more preferable range of the formula (2) is 34.2 to 48.0. The above upper and lower limits can each define a more preferable range of the formula (2).

[0069] The lower limit of formula (3) is preferably 1.11 or more, more preferably 1.22 or more, even more preferably 1.53 or more, still more preferably 1.63 or more, particularly preferably 1.68 or more, and most preferably 2.14 or more, 2.45 or more, 2.55 or more, 2.70 or more, 2.75 or more, 2.86 or more, 2.97 or more, or 3.06 or more. The upper limit of formula (3) is preferably 5.84 or less, more preferably 5.61 or less, even more preferably 5.10 or less, still more preferably 5.00 or less, particularly preferably 4.08 or less, and most preferably 3.98 or less, 3.88 or less, 3.60 or less, 3.57 or less, 3.52 or less, 3.47 or less, 3.37 or less, 3.26 or less, or 3.15 or less. A more preferable range of the formula (3) is 2.55 to 3.57. The above upper and lower limits can each define a more preferable range of the formula (3).

[0070] These formulas are obtained by the mutual dependence of each constituent element. This is because an alloy is a single entity formed by combining all of the constituent elements, and each constituent element influences the other. In this way, the solder alloy of the present invention, which adjusts the content of each constituent element to the optimum and further satisfies formulas (1) and (2), is set within a range that fully takes into account the mutual dependence of each constituent element. Therefore, when each constituent element is within the above-mentioned range and formulas (1) and (2) are further satisfied, various properties that have previously been difficult to achieve simultaneously can be achieved at a high level. Furthermore, when formula (3) is satisfied, various properties that have previously been difficult to achieve simultaneously can be achieved at a high level, as in the case of satisfying formulas (1) and (2).

[0071] In the examples described below, a rating of "◎" indicates that the result is particularly preferable in practical use compared to "◯." "◯" indicates a more preferable result than the prior art, and is therefore within the scope of the present invention, and is treated as an example. "X" indicates an insufficient result in the present invention, and is therefore outside the scope of the present invention, and is treated as a comparative example.

[0072] (11) Liftoff The solder alloy according to the present invention can suppress the occurrence of lift-off. Specifically, the solder paste according to the present invention, which will be described later, is paste-printed onto the electrodes of a six-layer printed circuit board (FR-4) that has been subjected to Cu-OSP treatment using a 150 μm-thick metal mask, the terminals of a four-terminal connector are inserted into the through-holes of the printed circuit board, and reflow is performed in a nitrogen atmosphere under heating conditions of holding at a temperature range of 220°C or higher for 40 seconds and achieving a peak temperature of 245°C, to prepare a test board, and when the cross section of the formed fillet is visually confirmed using an SEM photograph, it is preferable that there is no portion where the solder alloy is separated from the electrode.

[0073] (12) Heat cycle resistance (TCT) The solder alloy according to the present invention has excellent heat cycle resistance. That is, the solder alloy according to the present invention can suppress crack propagation in solder joints even after heat cycle testing. Specifically, a solder paste was prepared in the same manner as in the lift-off evaluation described above, and the paste was printed on the electrodes of a six-layer printed circuit board (FR-4) that had been subjected to Cu-OSP treatment using a 120 μm-thick metal mask. A 2012 chip resistor was then reflowed in a nitrogen atmosphere under heating conditions of 40 seconds at a temperature range of 220°C or higher, with a peak temperature of 245°C, to prepare a test substrate. This test substrate was subjected to 3,000 cycles under conditions of a low temperature of −40°C, a high temperature of +105°C, and a holding time of 30 minutes, after which it was removed from the heat cycle tester. The crack propagation rate of the solder joint was calculated using the following formula under the following conditions: It is acceptable if the resulting crack propagation rate is less than 90%, preferably 85% or less. The lower limit is acceptable if it is 0% or greater.

[0074] Crack growth rate (%) = (actual crack length) x 100 / (expected crack growth path length) The "expected crack propagation path length" refers to the predicted crack length in the event of complete fracture based on the actual crack measurement. If the actual crack is measured only near the bottom end of the chip resistor electrode, it is assumed that the crack reaches the fillet surface at a 45° angle from the fillet-side end of the actual crack, and the "expected crack propagation path length" is deemed to be the actual crack length and the distance from the end of the actual crack to the fillet surface. If the actual crack is measured only near the surface of the fillet and does not reach the bottom of the chip resistor electrode, the "expected crack propagation path length" is deemed to be the sum of the shortest distance from the chip resistor-side end of the actual crack to the electrode angle D on the Cu land side of the chip resistor electrode, the electrode length t from the electrode angle D to the chip resistor electrode, and the actual crack length.

[0075] (13) Wettability The solder alloy according to the present invention can have excellent wettability. In detail, after preparing a solder paste in the same manner as in the evaluation of lift-off described above, a solder paste having dimensions of 36 mm length × 50 mm width × 1.2 mm thickness and a solder resist opening of φ1.5 mm (area: 1.77 mm) was measured. 2 The substrate is a glass epoxy board (FR-4) with a Cu electrode of 0.24 mm diameter (area: 0.045 mm) and is OSP-treated. 2 Solder paste was printed onto the electrodes, which were the openings in the solder resist, through a metal mask measuring 1 / 4" x 0.1mm thick. The solder paste was then held at a temperature of 220°C or higher for 40 seconds, with a peak temperature of 245°C. Reflow was then carried out in a nitrogen atmosphere. The wetted and spread area was measured using a stereo microscope, and the wetted and spread area was found to be 0.08mm. 2 The solder resist opening was sufficiently larger than the area over which the solder alloy could spread, so the solder alloy did not reach the solder resist.

[0076] (14) Tensile strength The solder alloy according to the present invention can have high tensile strength. Specifically, the solder alloy according to the present invention is cast into a mold, and a test piece having a gauge length of 30 mm and a diameter of 8 mm is pulled at room temperature with an Instron Type 5966 at a stroke of 6 mm / min. When the test piece breaks, the tensile strength should be 80 MPa or more, preferably 85 MPa or more.

[0077] 2. Solder powder The solder powder according to the present invention is used in a solder paste, which will be described later, and is preferably a spherical powder. Being a spherical powder improves fluidity. The solder powder according to the present invention preferably satisfies the size (particle size distribution) of symbols 1 to 8 in the powder size classification (Table 2) of JIS Z 3284-1:2024. A size (particle size distribution) of symbols 4 to 8 is more preferable, and a size (particle size distribution) of symbols 5 to 8 is even more preferable. When the particle size satisfies this condition, the surface area of ​​the powder is not too large, suppressing an increase in viscosity. Furthermore, aggregation of the fine powder is suppressed, which may suppress an increase in viscosity. This enables soldering to finer components.

[0078] 3. Solder paste The solder paste according to the present invention is a mixture of a flux and the solder powder according to the present invention. The flux used in the present invention is not particularly limited as long as it allows soldering by conventional methods. Therefore, a suitable blend of commonly used rosin, organic acid, activator, and solvent may be used. The blending ratio of the solder powder component and the flux component is not particularly limited in the present invention, but is preferably 70 to 90 mass% of the solder powder component and 10 to 30 mass% of the flux component.

[0079] 4. Solder balls The solder alloy according to the present invention can be used as a solder ball. When used as a solder ball, the solder alloy according to the present invention can be manufactured using a dropping method, which is a common method in the industry. Alternatively, a solder joint can be manufactured by processing the solder ball using a common method in the industry, such as by mounting one solder ball on an electrode coated with flux and joining the solder balls. The particle size of the solder ball is preferably 1 μm or more, more preferably 10 μm or more, even more preferably 20 μm or more, and particularly preferably 30 μm or more. The upper limit of the particle size of the solder ball is preferably 3000 μm or less, more preferably 1000 μm or less, even more preferably 800 μm or less, and particularly preferably 600 μm or less.

[0080] 5. Solder preform The solder alloy according to the present invention can be used as a preform, which may be in the form of a washer, ring, pellet, disk, ribbon, wire, or the like.

[0081] 6. Solder joints The solder joint according to the present invention is suitable for use in joining at least two or more members to be joined. The members to be joined are not particularly limited, as long as they are electrically connected using the solder alloy according to the present invention, such as elements, substrates, electronic components, printed circuit boards, insulating substrates, heat sinks, lead frames, semiconductors using electrode terminals, power modules, inverter products, etc.

[0082] 7.Other The solder alloy of the present invention can be used in a conventional joining method, such as a reflow method. The melting temperature of the solder alloy when performing flow soldering is generally about 20°C higher than the liquidus temperature. Furthermore, when joining using the solder alloy of the present invention, the alloy structure can be further refined by taking into consideration the cooling rate during solidification. For example, the solder joint is cooled at a cooling rate of 2 to 3°C / s or more. Other joining conditions can be adjusted as appropriate depending on the alloy composition of the solder alloy.

[0083] The solder alloy according to the present invention can be manufactured by using a low-alpha ray material as its raw material, and when such a low-alpha ray alloy is used to form solder bumps around memory, it can suppress soft errors. [Example]

[0084] Using solder alloys having the alloy compositions shown in Tables 1 and 2, lift-off was evaluated as Evaluation 1, heat cycle resistance as Evaluation 2, wettability as Evaluation 3, and tensile strength as Evaluation 4. Each evaluation method is explained below.

[0085] Rating 1. Liftoff First, powders of each solder alloy listed in Tables 1 and 2 were prepared by atomization. Solder pastes were prepared by mixing these alloy powders with a flux ("GLV" manufactured by Senju Metal Industry Co., Ltd.) containing rosin, solvent, thixotropic agent, organic acid, etc. The alloy powder content of the solder paste was 88% by mass, and the flux content was 12% by mass. This solder paste was paste-printed onto the Cu lands of a six-layer printed circuit board (FR-4) that had been subjected to Cu-OSP treatment using a 150 μm-thick metal mask. Four four-terminal connectors were then prepared, and the terminals were inserted into the through-holes of the printed circuit board. The reflow process was carried out in a nitrogen atmosphere at a temperature of 220°C or higher for 40 seconds, with a peak temperature of 245°C. Test boards were then prepared.

[0086] Thirty-two fillets formed on the test substrate were cut, and the cross sections were photographed with an SEM. The SEM photographs were visually inspected to see if the fillets were separated from the Cu land. Figure 3 shows cross-sectional SEM photographs of the fillets, with Figure 3(a) being an SEM photograph enlarged 30 times, and Figure 3(b) being an SEM photograph enlarged 300 times. As shown in Figure 3(b), if there was even a small portion of the solder alloy separated from the Cu land in the cross section, lift-off had occurred, and the result was evaluated as "×." On the other hand, if there was no portion of the cross section separated from the Cu land, the result was evaluated as "Good."

[0087] Evaluation 2. Heat cycle resistance (TCT) Powders of each solder alloy listed in Tables 1 and 2 were prepared by atomization. Solder pastes were prepared by mixing these alloy powders with a flux (GLV, manufactured by Senju Metal Industry Co., Ltd.) containing rosin, solvent, thixotropic agent, organic acid, etc. The alloy powder content of the solder paste was 88% by mass, and the flux content was 12% by mass. This solder paste was printed onto the Cu lands of a six-layer printed circuit board (FR-4) that had been subjected to Cu-OSP treatment using a 120 μm-thick metal mask. Then, 2012 chip resistors were mounted using a mounter, and reflow was performed in a nitrogen atmosphere under heating conditions of 220°C or higher for 40 seconds, with a peak temperature of 245°C. Test boards were then prepared.

[0088] This test board was placed in a heat cycle tester set to low temperature of -40°C, high temperature of +105°C, and a holding time of 30 minutes. After 3000 cycles, the board was removed from the heat cycle tester under each condition, and the crack propagation rate of the cross section of the solder joint was measured.

[0089] The crack propagation rate was evaluated as follows. After 3000 cycles, 10 samples (20 fillets) were measured by cross-sectional observation using a FE-SEM JSM-7000F. The crack propagation rate was calculated using the following formula.

[0090] Crack growth rate (%) = (actual crack length) x 100 / (expected crack growth path length) Figure 4 shows the cracks 1 and 2 actually observed in a cross-sectional SEM photograph of a fillet, as well as the predicted crack propagation paths from these cracks. Here, the "actual crack length" refers to the total length of multiple cracks 1 and 2, as shown in Figure 4.

[0091] The "expected crack propagation path length" refers to the predicted crack length in the event of complete fracture based on the actual crack measurements. If the actual crack is measured only near the bottom end of the chip resistor electrode, it is assumed that the crack reaches the fillet surface at a 45° angle from the fillet-side end of the actual crack, and the "expected crack propagation path length" is considered to be the actual crack length and the distance from the end of the actual crack to the fillet surface. Also, if the actual crack is measured only near the surface of the fillet and does not reach the bottom of the chip resistor electrode, the "expected crack propagation path length" is considered to be the sum of the shortest distance from the chip resistor-side end of the actual crack to the electrode corner on the Cu land side of the chip resistor (D in the figure), the electrode length from electrode corner D to the chip resistor (t in Figure 6), and the actual crack length.

[0092] A crack propagation rate of 86% or more and less than 90% was judged as "Good", a rate of 85% or less was judged as "Excellent", and a rate of 90% or more was judged as "Poor".

[0093] Rating 3: Wettability Wetting and spreading tests were carried out in the following order, "1." and "2.", using solder pastes consisting of GLV-HF flux and solder powder and made from each solder alloy listed in Tables 1 and 2. The substrate used was a glass epoxy substrate (FR-4) measuring 36 mm long x 50 mm wide x 1.2 mm thick, equipped with a Cu electrode with a solder resist opening φ1.5 mm, and subjected to OSP treatment.

[0094] 1. Using the above substrate, electrodes were printed using a 0.24mm diameter x 0.1mm thick metal mask. Then, reflow was performed in a nitrogen atmosphere under heating conditions of 40 seconds at a temperature of 220°C or higher, with a peak temperature of 245°C, to produce an evaluation substrate.

[0095] 2. Using a stereo microscope, measure the wetted area and determine whether it is 0.08 mm 2 Spreading of more than 0.08mm was judged as "○". 2 Wetting and spreading of less than this was evaluated as "x".

[0096] Rating 4 Tensile strength The tensile strength was measured with reference to JIS Z3198-2:2003. Each solder alloy listed in Tables 1 and 2 was cast into a mold to prepare test pieces with a gauge length of 30 mm and a diameter of 8 mm. The test pieces were tensioned at room temperature with an Instron Type 5966 at a stroke of 6 mm / min, and the strength at which the test pieces broke was measured.

[0097] When the tensile strength was 85 MPa or more, it was evaluated as "A", when it was 80 MPa or more but less than 85 MPa, it was evaluated as "Good", and when it was less than 80 MPa, it was evaluated as "Poor".

[0098] The results are shown in Tables 1 and 2. Note that each solder alloy listed in Tables 1 and 2 contains unavoidable impurities. The amount of unavoidable impurities contained is very small, and these are contained in the remainder, Sn.

[0099] [Table 1]

[0100] [Table 2]

[0101] As shown in Tables 1 and 2, Examples 1 to 52 obtained results that were practically acceptable in all evaluations. Among these, all of the Examples that satisfied formulas (1) and (2) received the highest evaluation based on the above-mentioned criteria. Furthermore, it was found that all of the Examples that received the highest evaluation also satisfied formula (3).

[0102] On the other hand, Comparative Example 1 had poor TCT, wettability, and tensile strength due to the low Ag content, and Comparative Example 2 had poor TCT and tensile strength due to the high Ag content.

[0103] Comparative Examples 3 and 4 were inferior in TCT and tensile strength due to the inappropriate Cu content.

[0104] Comparative Examples 5 to 7 did not contain Bi or contained a small amount of Bi, and therefore at least exhibited poor wettability. Of these, Comparative Examples 5 and 6 exhibited poor TCT and tensile strength in addition to poor wettability. Comparative Examples 8 and 9 exhibited high Bi contents, and therefore lift-off occurred.

[0105] Comparative Examples 10 and 11 had low Sb contents, and therefore at least the TCT was poor. Of these, Comparative Example 10 also had poor tensile strength. Comparative Example 12 had a high Sb content, and therefore lift-off occurred, resulting in poor wettability.

[0106] In Comparative Examples 13 to 17, the Ni content was low, and therefore at least lift-off occurred. Of these, Comparative Example 13 also had poor tensile strength. Comparative Example 18 had poor wettability due to the high Ni content.

[0107] Comparative Examples 19 and 20 were alloy compositions containing Co, and even though the Sb content was 5.1%, the Sb content was too high, resulting in poor wettability. Comparative Examples 21 to 25 were alloy compositions containing Co, and even though the Ni content was 0.0050%, the Ni content was too low, resulting in lift-off. Comparative Example 26 was an alloy composition containing Co, and even though the Ni content was 0.200%, the Ni content was too high, resulting in poor wettability.

[0108] FIG. 5 shows SEM photographs of the cross section of a fillet magnified 300 times, with FIG. 5(a) being an SEM photograph of a fillet formed using Comparative Example 17, and FIG. 5(b) being an SEM photograph of a fillet formed using Example 3. Thus, it was found that in Example 3, the Cu land and the solder alloy were connected without any separation, and therefore lift-off did not occur. On the other hand, in Comparative Example 17, a portion of the Cu land and the solder alloy were separated, indicating that lift-off had occurred. Similar SEM photographs were also observed for the other Examples and Comparative Examples.

[0109] For the solder alloys listed in Tables 1 and 2, the liquidus and solidus temperatures during cooling were measured using the DSC curves during cooling. Then, the relationship between ΔT, the difference between the liquidus and solidus temperatures during cooling, and lift-off was investigated. The investigation method was as follows.

[0110] Evaluation 5. Solidus temperature, liquidus temperature, and ΔT during cooling The solidus temperature, liquidus temperature, and ΔT during cooling were determined from the DSC curves during cooling for the solder alloys shown in Tables 1 and 2. The DSC curves were obtained using a TA Instruments DSC (model number: Q2000) by heating the alloy in air at a heating rate of 5°C / min to 240°C, a temperature above the liquidus temperature, and then cooling it at a heating rate of 60°C / min to 25°C.

[0111] From the DSC curves thus obtained, graphs were created in which the horizontal axis represents temperature and the vertical axis represents the solid fraction, as shown in Fig. 6. Fig. 6 is a graph showing the relationship between temperature and solid fraction obtained from the DSC curves during cooling, with Fig. 6(a) being the graph for Example 3 and Fig. 6(b) being the graph for Comparative Example 13.

[0112] The solid fraction was calculated by numerically integrating (trapezoidal integration) the distance between the DSC curve and the baseline for each time (second) during cooling. Specifically, the time when the heat quantity began to increase from the baseline was defined as ts seconds (0 seconds), and the time when the heat quantity returned to the baseline was defined as tf seconds. The heat quantity from ts to ts+1 seconds was then integrated, followed by the heat quantity from ts+1 to ts+2 seconds, and then the heat quantity for each second was integrated sequentially. Finally, the heat quantity from ts+x to tf seconds was integrated. The solid fraction was then calculated by dividing the total integral value from ts to tx seconds by the total integral value from ts to tf and multiplying the result by 100%. Furthermore, when obtaining the DSC curve, the sample temperature was also measured for each time (second) during cooling, and the horizontal axis was changed from time to temperature.

[0113] The calculated solid fraction was plotted for each temperature during cooling, with the vertical axis representing the calculated solid fraction and the horizontal axis representing the temperature during cooling. A graph was then obtained by connecting each plot with a straight line and smoothing. In Figure 6, the temperature at which the solid fraction was 0% was determined as the liquidus temperature, and the temperature at which the solid fraction was 100% was determined as the solidus temperature. The range of the numerical integration described above was determined to be from the start point of solidification to the end point of solidification.

[0114] Comparing Figures 6(a) and 6(b), it is clear that Figure 6(a), which is the graph for Example 3, has a narrower ΔT, which is the temperature difference between the solid fraction of 0% and 100%, compared to Figure 6(b), which is the graph for Comparative Example 14. Similarly, ΔT was determined for each of the other solder alloys listed in Tables 1 and 2. From these, Examples and Comparative Examples that demonstrate the criticality of lift-off are extracted in Tables 3 to 5, and it is shown that for the extracted alloy compositions, a narrow ΔT suppresses the occurrence of lift-off.

[0115] Table 3 is a table extracting Example 15, in which the Bi content was the upper limit and the lift-off evaluation was "good," and Comparative Example 8, in which the Bi content exceeded the upper limit and the lift-off evaluation was "poor." Table 4 is a table extracting Example 24, in which the Sb content was the upper limit and the lift-off evaluation was "good," and Comparative Example 12, in which the Sb content exceeded the upper limit and the lift-off evaluation was "poor." Table 5 is a table extracting Example 25, in which the Ni content was the lower limit and the lift-off evaluation was "good," and Comparative Example 13, in which the Ni content was less than the lower limit and the lift-off evaluation was "poor."

[0116] [Table 3]

[0117] [Table 4]

[0118] [Table 5]

[0119] As is clear from Tables 3 to 5, for Bi, Sb, and Ni, which show the criticality of lift-off, the examples in which the lift-off was evaluated as "Good" all had narrower ΔT than the comparative examples in which the lift-off was evaluated as "Poor."

Claims

1. A solder alloy characterized by having an alloy composition consisting of, in mass%, 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.5% Sb, 0.050 to 0.200% Ni, and the balance being Sn.

2. A solder alloy characterized by having an alloy composition consisting of, in mass%, 3.0 to 4.0% Ag, 0.60 to 0.80% Cu, 0.8 to 2.3% Bi, 1.2 to 5.0% Sb, 0.060 to 0.190% Ni, 0.001 to 0.3% Co, and the balance being Sn.

3. 2. The solder alloy according to claim 1, wherein the alloy composition further contains, in mass%, 0.11% or less in total of at least one of Ga, As, Fe, Pd, Mn, In, Zn, Zr, and Mg.

4. The solder alloy according to claim 1 or 2, wherein the alloy composition satisfies the following formulas (1) and (2): 0.216≦(Ag×Cu) / (Bi×Sb)≦0.465 (1) 34.2≦Bi×Sb / Ni≦85.6 (2) In the above formulas (1) and (2), Ag, Cu, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

5. The solder alloy according to claim 4 , wherein the alloy composition satisfies the following formula (3): 1.11≦Ag×Bi×Sb×Ni≦5.84 (3) In the above formula (3), Ag, Bi, Sb, and Ni each represent the content in mass % of the alloy composition.

6. A solder powder comprising the solder alloy according to claim 1 or 2.

7. A solder powder comprising the solder alloy according to claim 3.

8. A solder paste comprising the solder powder of claim 6.

9. A solder paste comprising the solder powder of claim 7.

10. A solder ball made of the solder alloy according to claim 1 or 2.

11. A solder ball made of the solder alloy according to claim 3.

12. A solder preform made of the solder alloy according to claim 1 or 2.

13. A solder preform made of the solder alloy according to claim 3.

14. A solder joint comprising the solder alloy of claim 1 or 2.

15. A solder joint comprising the solder alloy of claim 3.

16. A solder alloy as described in claim 2, wherein the alloy composition further contains, by mass %, at least one of Ge, Ga, As, Fe, Pd, Mn, In, Zn, Zr, and Mg in a total amount of 0.11% or less.

17. A solder powder comprising the solder alloy described in claim 16.

18. A solder paste comprising the solder powder of claim 17.

19. A solder ball made of the solder alloy of claim 16.

20. A solder preform comprising the solder alloy of claim 16.

21. A solder joint having the solder alloy of claim 16.

Images