WAFER PROCESSING METHODS

Abstract

A wafer processing method for processing a wafer (11) having components (15) formed in regions on one side of a front surface (11a) of the wafer (11), the regions being defined by first planned separation lines (13a) and second planned separation lines (13b) intersecting the first planned separation lines (13a), the wafer processing method comprising: a protective element bonding step comprising bonding a protective element (17) to the front surface (11a) side of the wafer (11), a first modified layer forming step comprising irradiating a laser beam (8), the laser beam (8) having a transmissivity through the wafer (11), from a rear surface (11b) side of the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), with a focal point of the laser beam (8) set at a height that is at a height of a first region (11c) arranged inside the wafer (11), the wafer (11) being held via the protective member (17) by a chuck table (4) on the front surface (11a) side of the wafer (11), thereby forming first modified layers (23a) in the first region (11c), after performing the first modified layer forming step, a second modified layer forming step comprising irradiating the laser beam (8) from the rear surface (11b) side of the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), with the focal point of the laser beam (8) set at a height that lies at a height of a second region (11d) arranged inside the wafer (11) on the rear surface (11b) side of the wafer (11), whereby second modified layers (23b) are formed in the second region (11d) and cracks (25) are formed that extend from the front surface (11a) to the rear surface (11b) of the wafer (11) to separate the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), and after performing the second modified layer forming step, a grinding step comprising grinding the wafer (11) on the back surface (11b) side of the wafer (11) to thin the wafer (11) to a predetermined thickness.

Description

BACKGROUND OF THE INVENTION FIELD OF THE INVENTION

[0001] The present invention relates to a processing method for a wafer with components formed thereon. DESCRIPTION OF THE RELATED PRIOR ART

[0002] In a component chip manufacturing step, a wafer is used with components, such as integrated circuits (ICs) or large-scale integrations (LSIs), each formed in a plurality of regions. This plurality of regions is defined by a plurality of planned separation lines (streets) and another plurality of planned separation lines that intersect each other. A plurality of component chips, each comprising the components, are obtained by dividing the wafer along each planned separation line.

[0003] For example, a cutting device is used to separate the wafer. The cutting device includes a spindle (shaft) with attached annular cutting blades for cutting the wafer. The cutting blades are rotated and caused to cut into the wafer along the planned separation lines, thereby cutting and separating the wafer along the planned separation lines.

[0004] On the other hand, in recent years, interest has focused on technologies that separate wafers by laser processing. For example, a method was proposed in Japanese Patent Laid-Open No. 2011-49454. According to this method, a laser beam having transmissivity through a wafer is focused on the inside of the wafer to form modified regions (modified layers) along planned separation lines inside the wafer. The regions with the modified layers formed therein are made more brittle than the remaining regions in the wafer. Therefore, when an external force is applied to the wafer having the modified layers formed therein, the wafer is separated along the planned separation lines with the modified layers as starting points.

[0005] For example, a film (stretchable film) that can be stretched by applying an external force is adhered to a wafer with modified layers formed therein. By causing the film to stretch, an external force is applied to the wafer, so that the wafer is separated along the intended separation lines. Furthermore, a method was proposed in Japanese Patent Application Laid-Open No. JP 2015-37172 A. According to this method, a wafer with modified layers formed therein is subjected to grinding, thereby generating cracks from the modified layers to separate the wafer.

[0006] Further prior art helpful for understanding the present invention can be found in the following document: JP 2014-78 556 A relates to a method for successfully dividing a wafer without causing a crack or chipping on a divided chip. SUMMARY OF THE INVENTION

[0007] When dicing a wafer by laser processing as described above, a plurality of modified layers can be formed in a wafer thickness direction along individual planned dicing lines depending on the wafer thickness, material, and / or the like. These plurality of modified layers are formed, for example, by irradiating a laser beam from an upper surface side (a back surface side) of the wafer along each individual planned dicing line a plurality of times while gradually changing a position (height) of a focal point of the laser beam from the lower surface side toward the upper surface side of the wafer. In this way, the wafer is properly separated from the modified layers as starting points even when the wafer is relatively thick, for example.

[0008] When forming a plurality of modified layers along each of the first planned division lines and the second planned division lines that cross each other, it is now common practice to form a plurality of modified layers from the lower surface side to the upper surface side of a wafer along the first planned division lines, and then to form a plurality of modified layers from the lower surface side to the upper surface side of the wafer along the second planned division lines. Therefore, when forming the modified layers along the second planned division lines, the former plurality of modified layers are already formed from the lower surface side to the upper surface side of the wafer in intersecting regions of the first planned division lines and the second planned division lines.

[0009] When forming the latter modified layers along the second planned separation lines, the laser beam is also irradiated onto the intersecting regions. The laser beam irradiated onto the former modified layers already formed in the intersecting regions is subject to diffuse scattering (dispersion), so the laser beam may be deflected by the second planned separation lines. As a result, the laser beam may also be irradiated onto devices formed on the lower surface side (front surface side) of the wafer, potentially causing device damage.In particular, when forming the latter modified layers on the lower surface side of the wafer along the second planned separation lines, the laser beam is also irradiated onto the plurality of former modified layers already formed in the intersecting regions from the lower surface to the upper surface of the wafer. Accordingly, diffuse scattering of the laser beam tends to occur, making the devices vulnerable to damage.

[0010] In view of the above problem, it was an object of the present invention to provide a wafer processing method that counteracts damage to the components that occurs due to irradiation with a laser beam.

[0011] In accordance with one aspect of the present invention, a wafer processing method is provided for processing a wafer having components formed in regions on a front surface side of the wafer. The regions are defined by first planned dividing lines and second planned dividing lines that cross the first planned dividing lines. The processing method includes a protective element bonding step of bonding a protective element to the front surface side of the wafer, a first modified layer formation step of irradiating a laser beam, the laser beam having a transmissivity for the wafer, from a rear surface side of the wafer along the first planned dividing lines and the second planned dividing lines with a focal point of the laser beam set at a height equal to a height of a first region,which is arranged inside the wafer, wherein the wafer is held via the protective member by a chuck table on the front surface side of the wafer, whereby first modified layers are formed in the first region; a second modified layer formation step after performing the first modified layer formation step by irradiating the laser beam from the back surface side of the wafer along the first planned division lines and the second planned division lines with the focal point of the laser beam set at a height equal to a height of a second region arranged inside the wafer on the back surface side of the wafer, whereby second modified layers are formed in the second region and also cracks are formed extending from the front surface to the back surface of the wafer;to separate the wafer along the first planned separation lines and the second planned separation lines, and a grinding step after performing the second modified layer forming step with grinding the wafer on the back surface side of the wafer to thin the wafer to a predetermined thickness.,

[0012] In the first modified layer forming step, cracks may preferably be formed extending from the first modified layers to the front surface of the wafer. Further, in the first modified layer forming step, the first modified layers may be formed such that the front surface of the wafer and the first modified layers have a distance greater than the predetermined thickness therebetween, and in the grinding step, the wafer may preferably be thinned to the predetermined thickness, thereby removing the first modified layers.

[0013] In the wafer processing method according to the first aspect of the present invention, the first modified layers are first formed along the first planned division lines and the second planned division lines in the first region located inside the wafer on the front surface side (the lower surface side) of the wafer. The second modified layers are then formed along the first planned division lines and the second planned division lines in the second region located inside the wafer on the back surface side (the upper surface side) of the wafer. According to the wafer processing method described above, when the first modified layers are formed in the first region along the first planned division lines and the second planned division lines, no modified layers are yet formed in the second region.Consequently, virtually no diffuse scattering (distribution) of the laser beam occurs in the second region when the laser beam is irradiated onto the wafer with its focal point positioned in the first region. As a result, there is essentially no irradiation of the components with the laser beam, thus preventing damage to the components.

[0014] The above and other objects, features and advantages of the present invention and the mode for carrying them out will become more apparent and the invention itself will be best understood by studying the following description and appended claims with reference to the accompanying drawings which show a preferred embodiment of the invention. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 is a perspective view showing a wafer; Fig. 2 is a partially sectional side view illustrating a laser processing apparatus; Fig. 3A is a sectional view illustrating an enlarged portion of the wafer during a first modified layer forming step; Fig. 3B is a sectional view illustrating, in an enlarged manner, the portion of the wafer after the first modified layer forming step; Fig. 4A is a sectional view illustrating the portion of the wafer during a step of forming a second modified layer; Fig. 4B is a sectional view illustrating the portion of the wafer after the second modified layer forming step on an enlarged scale; Fig. 5 is a side view illustrating a grinding device; and Fig. 6 is a sectional view illustrating a portion of the wafer after a grinding step on an enlarged scale. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENT

[0015] With reference to the accompanying drawings, an embodiment according to one aspect of the present invention will be described below. First, an embodiment of a wafer that can be processed by the wafer processing method according to the present embodiment will be described. Fig. 1 is a perspective view illustrating a wafer 11.

[0016] The wafer 11 is formed into a disk shape using a material such as silicon and has a front surface 11a and a back surface 11b. The wafer is defined as a plurality of regions by a plurality of planned dividing lines (streets) and another plurality of planned dividing lines (streets) arranged in a lattice structure so as to cross each other. Specifically, first planned dividing lines 13a and second planned dividing lines 13b established on the wafer 11 will be described in detail. The first planned dividing lines 13a are aligned so that their longitudinal directions extend in a first direction (a direction indicated by arrow A), and the second planned dividing lines 13b are aligned so that their longitudinal directions extend in a second direction (a direction indicated by arrow B) substantially perpendicular to the first direction.The first planned dividing lines 13a and the second planned dividing lines 13b are arranged to cross each other. The wafer 11 is divided into a plurality of regions by the first planned dividing lines 13a and the second planned dividing lines 13b. On one side of the front surface 11a of the wafer 11, components 15 configured as ICs, LSIs, or the like are formed in the respective regions defined by the first planned dividing lines 13a and the second planned dividing lines 13b. When the wafer 11 is divided along the first planned dividing lines 13a and the second planned dividing lines 13b, a plurality of component chips each having the components 15 are obtained.

[0017] There are no restrictions on the material, shape, structure, size, and the like of the wafer 11. For example, the wafer 11 may also be formed of a material other than silicon, such as a semiconductor (GaAs, SiC, InP, GaN, or the like), sapphire, glass, ceramic, resin, or metal. Furthermore, there are no restrictions on the type, number, shape, structure, size, layout, and the like of the components 15.

[0018] When dicing the wafer 11, for example, dicing start points are formed along the first planned dicing lines 13a and the second planned dicing lines 13b. These dicing start points serve as dicing start points (dicing triggers) when dicing the wafer 11. The dicing start points are formed, for example, by applying laser processing to the wafer 11 and modifying (or influencing) an interior of the wafer 11 along the first planned dicing lines 13a and the second planned dicing lines 13b. Modified regions (modified layers) in the interior of the wafer 11 are designed to be more brittle than the remaining regions of the wafer 11.Then, when an external force directed outward in a radial direction of the wafer 11, for example, is applied to the wafer 11 with the modified layers formed therein, cracks propagate from the modified layers in the thickness direction of the wafer 11, so that the wafer 11 is separated from the modified layers as starting points. In other words, the modified layers serve as separation starting points for the wafer 11. As a result, the wafer 11 is separated along the first planned separation lines 13a and the second planned separation lines 13b, and the plurality of device chips including the devices 15 are obtained, respectively.

[0019] When forming separation start points in the wafer 11 by laser processing, first, a protective member 17 is bonded to the front surface 11a side of the wafer 11 (protective member bonding step). As the protective member 17, for example, a film-shaped tape made of a resin or the like and formed into a circular shape with substantially the same radius as the wafer 11 is used. The components 15 formed on the front surface 11a side of the wafer 11 are covered and protected by the protective member 17. For easy processing and transfer of the wafer 11, the wafer 11 may be supported by an annular frame.Specifically, a protective member 17 having a circular shape, a radius larger than that of the wafer 11, and bonded to the front surface 11a side of the wafer 11 will be described, and an annular frame having a circular opening with a radius larger than that of the wafer 11 is bonded to an outer peripheral portion of the protective member 17. As a result, a frame unit is configured with the wafer 11 supported by the annular frame via the protective member 17.

[0020] Next, the wafer 11 is held by a laser processing device. Fig. 2 is a partial sectional side view illustrating a laser processing apparatus 2. The laser processing apparatus 2 includes a chuck table (holding table) 4 that holds the wafer 11 and a laser irradiation unit 6 that irradiates a laser beam 8.

[0021] The chuck table 4 is connected to an unillustrated rotation drive source such as an electric motor, and this rotation drive source rotates the chuck table 4 around a rotation axis substantially parallel to a vertical direction (Z-axis direction, up-down direction). Further, an unillustrated moving mechanism is arranged below the chuck table 4, and this moving mechanism moves the chuck table 4 in a processing feed direction (X-axis direction, first horizontal direction) and in a dividing direction (Y-axis direction, second horizontal direction). The chuck table 4 has an upper surface constituting a holding surface 4a that holds the wafer 11. According to the shape of the wafer 11, the holding surface 4a is formed in a circular shape. However, the shape of the holding surface 4a can be appropriately changed depending on the shape of the wafer 11 or the like.The holding surface 4a is connected to a suction source (not shown) via a flow channel (not shown) formed inside the clamping table 4.

[0022] The laser irradiation unit 6 is arranged above the chuck table 4. The laser irradiation unit 6 includes a laser oscillator that outputs a laser beam having a predetermined wavelength in pulse waveform, and a condenser that focuses the laser beam output from the laser oscillator to a predetermined position. The laser irradiation unit 6 irradiates the laser beam 8 toward the wafer 11 held by the chuck table 4. The wavelength of the laser beam 8 is adjusted so that the laser beam 8 has a transmissivity through the wafer 11. Accordingly, the laser irradiation unit 6 irradiates a laser beam 8 that is transmissible through the wafer (i.e., has a transmissivity for the wafer 11) onto the wafer 11.

[0023] The rotation drive source and the movement mechanism connected to the chuck table 4 and the laser irradiation unit 6 are each connected to a control section (not shown) configured to control individual elements of the laser processing apparatus 2. The position of the chuck table 4 and the irradiation conditions (focal point position, power, spot diameter, repetition frequency, etc.) of the laser beam 8 are controlled by the control section.

[0024] The wafer 11 is held by the chuck table 4 via the protective member 17. Specifically, it is described that the wafer 11 is placed on the chuck table 4 such that the front surface 11a side of the wafer 11 (i.e., the side of the protective member 17) and the holding surface 4a of the chuck table 4 face each other. When a negative pressure of the suction source is applied to the holding surface 4a in this state, the wafer 11 is held by the chuck table 4 via the negative pressure. Next, the laser beam 8 is irradiated from the laser irradiation unit 6 toward the wafer 11. At this time, the laser beam 8 is focused on the inside of the wafer 11 (in a region between the front surface 11a and the back surface 11b of the wafer). Further, the irradiation conditions (the power, the spot diameter, the repetition frequency, etc.) are set.) of the laser beam 8 is adjusted so that the interior of the wafer 11 is modified (affected) by multiphoton absorption. By moving the chuck table 4 in the horizontal direction (the X-axis direction in . Fig. 2) During irradiation of the laser beam 8 onto the wafer 11, a modified linear layer (affected linear layer) 19 (hereinafter referred to simply as "the modified layer (affected layer) 19") is formed inside the wafer 11. The region with the modified layer 19 formed therein is more brittle than the remaining regions of the wafer 11. Then, for example, when an external force is applied to the wafer 11 with the modified layer 19 formed therein, cracks propagate from the modified layer 19 in the thickness direction of the wafer 11, so that the wafer 11 is separated from the modified layer 19 as starting points. In other words, the modified layer 19 serves as separation starting points.

[0025] Note that, depending on the thickness, material, and / or the like of the wafer 11, a plurality of modified layers 19 may be formed in the thickness direction of the wafer 11. By forming such a plurality of modified layers 19, the wafer 11 can be properly separated from the modified layers 19 as starting points even if the wafer 11 is relatively thick, for example.

[0026] When forming such a plurality of modified layers 19 in the thickness direction of the wafer 11, it is customary to form a separation layer between the lower surface side (the front surface side 11a) and the upper surface side (the rear surface side 11b) of the wafer 11 along the first planned separation lines 13a (see Fig. 1) to form a plurality of modified layers 19 and then between the lower surface side to the upper surface side of the wafer along the second planned separation lines 13b (see Fig. 1) to form a plurality of additional modified layers 19. When forming the additional modified layers 19 along the second planned separation lines 13b, the first plurality of modified layers 19 are thus already sandwiched between the side of the lower surface and the side of the upper surface of the wafer 11 in regions 13c (see Fig. 1) of the first planned separation lines 13a and the second planned separation lines 13b, which cross each other, are formed. When forming the latter modified layers 19 along the second planned separation lines 13b, the laser beam 8 is also irradiated onto the intersecting regions 13c. Therefore, the laser beam 8 is also irradiated onto the former modified layers 19 already formed in the intersecting regions 13c. Consequently, diffuse scattering (distribution) of the laser beam 8 is caused, so that the laser beam 8 may deviate from the second planned separation lines 13b. As a result, the laser beam 8 may also be irradiated onto the components 15 (see Fig. 1) formed on the lower surface side (the front surface 11a side) of the wafer 11, possibly resulting in damage to the devices 15. Specifically, when forming the latter modified layers 19 on the lower surface side (the front surface 11a side) of the wafer 11 along the second planned separation lines 13b, the laser beam is also irradiated onto the former modified layers 19 already formed in the intersecting regions 13c between the lower surface and the upper surface of the wafer 11. Accordingly, diffuse scattering of the laser beam 8 tends to occur, making the devices 15 vulnerable to damage.

[0027] Therefore, in the wafer processing method according to the present embodiment, first, two modified layers are formed in a first region 11c (see Fig. 3A, etc.) arranged inside the wafer 11 on the side of the front surface 11a of the wafer 11, along the first planned separation lines 13a and the second planned separation lines 13b, and then two additional modified layers are formed in a second region 11d (see Fig. 4A, etc.) arranged inside the wafer 11 on the side of the rear surface 11b of the wafer 11, is formed along the first planned dividing lines 13a and the second planned dividing lines 13b.

[0028] Specifically, it is described that two first modified linear layers (two first influenced linear layers) 23a (hereinafter referred to simply as “the first modified layers (first influenced layers) 23a”) are first formed in the first region 11c of the wafer 11 along the planned separation lines 13a and the second planned separation lines 13b (first modified layer forming step). Fig. 3A is a sectional view illustrating, on an enlarged scale, a portion of the wafer 11 during the formation step of a first modified layer. On the front surface 11a side of the wafer 11, the protective element 17 is connected via a component layer (functional layer) 21. The component layer 21 corresponds to a layer including various functional layers (conductive layer, insulating layer, and the like) associated with the components 15 (see Fig. 1) which are formed on the side of the front surface 11a of the wafer 11.

[0029] During the formation step of a first modified layer, the clamping table 4 (see Fig. 2) first rotated to align the longitudinal direction of one of the first planned dividing lines 13a (see Fig. 1) to align with the processing feed direction of the chuck table 4. Furthermore, the height of the focal point of the laser beam 8 in the first region 11c of the wafer 11 is set to a predetermined first height. Then, the chuck table 4 is moved in the processing feed direction during irradiation of the laser beam 8 from the laser irradiation unit 6 toward the wafer 11, whereby the laser beam 8 is irradiated along the first planned separation line 13a. During the irradiation of the laser beam 8, one of the first modified layers 23a, which is in Fig. 3A and hereinafter referred to as "the lower first modified layer 23a" is formed at the predetermined first height in the first region 11c of the wafer 11 along the one first planned division line 13a, and further, cracks 25a propagate from the lower first modified layer 23a in the thickness direction of the wafer 11. Note that the cracks 25a may be formed from the lower first modified layer 23a toward the front surface 11a or the back surface 11b, or may be formed from the lower first modified layer 23a toward both the front surface 11a and the back surface 11b.

[0030] Next, the focal point of the laser beam 8 is moved toward the rear surface 11b in the first region 11c of the wafer 11, and the laser beam 8 is irradiated onto the first region 11c in a similar manner (see Fig. 3A). As a result, the other first modified layer 23a formed on a layer formed in the Fig. 3A and Fig. 3B and hereinafter referred to as "the upper first modified layer 23a," is formed at a second predetermined height that is higher than the first predetermined height in the first region 11c, whereby the two first modified layers 23a, which overlap each other when viewed from above, are formed along the one first planned dividing line 13a. It should be noted that an example in which the first modified layers 23a are formed up to twice in Fig. 3A. However, the number of first modified layers 23a can be changed depending on the thickness, material, and / or the like of the wafer 11, and can be set to a desired positive integer of 1 or more. When a plurality of first modified layers 23a are formed, cracks 25a propagating from one of the first modified layers 23a and cracks 25a propagating from one or more remaining first modified layers 23a are connected to each other. Fig. 3A, the cracks 25a propagating from the lower first modified layer 23a (the first modified layer 23a on the front surface 11a side) toward the rear surface 11b side and the cracks 25a propagating from the upper first modified layer 23a (the first modified layer 23a on the rear surface 11b side) toward the front surface 11a are connected to each other.

[0031] Additional first modified layers 23a are then formed along the remaining first planned separation lines 13a in a similar manner and sequentially, with up to two layers per first planned separation line 13a. As a result, the first modified layers 23a are formed in the first region 11c of the wafer 11 up to twice along each first planned separation line 13a.

[0032] Next, the clamping table 4 is rotated to align the longitudinal direction of one of the second planned parting lines 13b (see Fig. 1) to align with the processing feed direction of the chuck table 4. Following a similar procedure, first modified layers 23a are formed up to twice per second planned separation line 13b along all of the second planned separation lines 13b. As a result, the first modified layers 23a are formed up to twice, respectively, in a lattice pattern at the predetermined first and second heights in the first region 11c of the wafer 11 along the first planned separation lines 13a and the second planned separation lines 13b.

[0033] When forming the first modified layer 23a along every second planned dividing line 13b, the laser beam 8 is radiated through the second region 11d onto the first region 11c. If, at this time, modified layers are already formed in the second region 11d, the laser beam 8 may be radiated onto the modified layers and subjected to diffuse scattering (distribution), and may then be radiated onto at least some of the components 15 belonging to the component layer 21. As a result, such components 15 may possibly be damaged. In the first modified layer formation step described above, on the other hand, the first modified layer 23a is formed only in the first region 11c along every first planned dividing line 13a.Accordingly, when forming the first modified layer 23a along every second planned separation line 13b, no modified layers are yet formed in the second region 11d, so that the laser beam 8 experiences virtually no diffuse scattering in the second region 11d. As a result, the laser beam 8 is hardly radiated onto the components 15, thus preventing damage to the components 15.

[0034] When forming a plurality of first modified layers 23a along each of the first planned parting lines 13a and the second planned parting lines 13b, there is no limitation on the order in which the first modified layers 23a are formed. For example, first, as many first modified layers 23a as a plurality per first planned parting line 13a may be formed along the first planned parting lines 13a, and then, as many first modified layers 23a as a plurality per second planned parting line 13b may be formed along the second planned parting lines 13b.Alternatively, the first modified layers 23a may be formed one after another along the first planned parting lines 13a and the second planned parting lines 13b, followed by repeating the previous step with the focal point of the laser beam 8 set at a height closer to the rear surface 11b side.

[0035] Fig. Fig. 3B is a sectional view illustrating, on an enlarged scale, the portion of the wafer 11 after the first modified layer formation step. In the first region 11c of the wafer 11, two first modified layers 23a are formed in the longitudinal direction of each first planned separation line 13a (the left-right direction of the side plane of the Fig. 3B) and additional two first modified layers 23a are formed in the longitudinal direction of every second planned parting line 13b (the front-rear direction of the side plane of the Fig. 3B) are trained. Fig. 3B illustrates an example of the wafer 11 in which the first modified layers 23a are formed along the first planned separation lines 13a and the second planned separation lines 13b up to twice per planned separation line.

[0036] When the first modified layers 23a are formed, the regions of the wafer 11 to which the first modified layers 23a belong may expand, so that warpage may occur in the wafer 11 due to the expansion. When such warpage occurs in the wafer 11, the positions of the first planned dividing lines 13a and the second planned dividing lines 13b change, making it difficult to irradiate the laser beam 8 along the first planned dividing lines 13a and the second planned dividing lines 13b. In particular, when the first planned dividing lines 13a and the second planned dividing lines 13b are numerous and their pitches are small (for example, 5 mm or less), the positions of the first planned dividing lines 13a and the second planned dividing lines 13b are susceptible to large changes due to the expansion of the wafer 11.In the first modified layer forming step, therefore, it is preferable to cause the cracks 25a to be formed as shown in . Fig. 3A illustrates, cracks occur from the first modified layers 23a to the front surface 11a of the wafer 11. Specifically, when forming each lower first modified layer 23a closer to the front surface 11a of the wafer 11, the irradiation conditions of the laser beam 8 are adjusted so that cracks 25a are formed that extend from the lower first modified layer 23a and reach the front surface 11a of the wafer 11. For example, when a silicon wafer with a diameter of 12 inches and a thickness of 775 μm is used as the wafer 11, the focal point of the laser beam 8 is positioned at a point in terms of the distance (depth) from the front surface 11a of the wafer 11 that is 200 μm or less deep. The laser beam 8 is then emitted along the first planned dividing line 13a and the second planned dividing line 13b.The irradiation conditions of the laser beam 8 can be set, for example, as follows:. Light source: YAG laser Wavelength: 1064 nm Repetition frequency: 60 kHz Average output power: 1.8 watts Machining feed speed: 900 mm / s

[0037] When the laser beam 8 is irradiated under the conditions described above, the cracks 25a are continuously formed, extending from the lower first modified layer 23a, which is closer to the front surface 11a of the wafer 11, and reaching the front surface 11a of the wafer 11. As a result, the wafer 11 is formed in some areas on the front surface 11a side along the corresponding first planned division lines 13a and / or the corresponding second planned division lines 13b. It has been confirmed that when the wafer is divided on the front surface 11a side as described above, the wafer 11 resists the occurrence of warpage even when the first modified layers 23a are formed.As a result, distortion of the first planned dividing lines 13a and the second planned dividing lines 13b is suppressed, thereby facilitating irradiation of the laser beam 8 along the first planned dividing lines 13a and the second planned dividing lines 13b. Furthermore, at the time the first modified layers 23a have been formed, the wafer 11 continues to remain intact without division on the rear surface 11b side (the second region 11d side). Therefore, even if expansion occurs on the front surface 11a side (the first region 11c side) of the wafer 11 due to the formation of the first modified layers 23a, the wafer 11 as a whole is hardly subject to distortion or warpage, thus counteracting positional changes of the first planned dividing lines 13a and the second planned dividing lines 13b.

[0038] Next, second modified layers (affected layers) 23b are formed in the second region 11d of the wafer 11 along the first planned separation lines 13a and the second planned separation lines 13b (second modified layer formation step). Fig. 4A is a sectional view illustrating the portion of the wafer 11 during the step of forming a second modified layer on an enlarged scale.

[0039] In the second modified layer forming step, the laser beam 8 is irradiated along the first planned separation lines 13a and the second planned separation lines 13b by similar operations as in the first modified layer forming step. However, when irradiating the laser beam 8 onto the wafer 11, the height of the focal point of the laser beam 8 in the second region 11d of the wafer 11 is set to a predetermined first height. As a result, the second modified linear layers 23b formed on a Fig. 4A and hereinafter referred to as "the lower second modified layers 23b" are formed at the predetermined first height in the second region 11d of the wafer 11 along the first planned division lines 13a and the second planned division lines 13b, and cracks 25b are also formed from the lower second modified layers 23b along the thickness direction of the wafer 11. The above-described irradiation step is next repeated except that the height of the focal point of the laser beam 8 is aligned at the predetermined second height on a side closer to the rear surface 11b than the predetermined first height, that is, in other words, at a height in Fig. 4A. As a result, the additional second modified linear layers 23b, hereinafter referred to as “the upper second modified layers 23b,” are formed at the predetermined second height in the second region 11d of the wafer 11 along the planned dividing lines 13a and the second planned dividing lines 13b, and further, cracks 25b are formed from the upper second modified layers 23b in the thickness direction of the wafer 11. In the second modified layer forming step, the irradiation conditions of the laser beam 8 are set such that, as shown in Fig. 4A, cracks 25b are formed extending from the upper second modified layers 23b and reaching the rear surface 11b of the wafer 11.

[0040] Fig. 4B is a sectional view illustrating the portion of the wafer 11 after the second modified layer forming step on an enlarged scale. Fig. 4B illustrates an example of the wafer 11 in which the second modified layers 23b are formed along the first planned division lines 13a and the second planned division lines 13b up to twice per planned division line. Furthermore, the cracks 25b propagating from the lower second modified layer 23b (the second modified layer 23b on the front surface 11a side) toward the back surface 11b side and the cracks 25b propagating from the upper second modified layer 23b (the second modified layer 23b on the back surface 11b side) toward the front surface 11a are connected to each other. Furthermore, the number of second modified layers 23b to be formed in the second region 11d can be appropriately set like the number of first modified layers 23a.When forming a plurality of second modified layers 23b along each of the first planned parting lines 13a and the second planned parting lines 13b, the order in which the second modified layers 23b are to be formed can be appropriately adjusted, as in the first modified layer forming step. Furthermore, the cracks 25b propagate in a similar manner to the cracks 25a formed in the first region 11c.

[0041] When the second modified layer forming step is performed, the cracks 25a formed in the first modified layer forming step and the cracks 25b formed in the second modified layer forming step are connected to each other, whereby a plurality of cracks 27 extending from the rear surface 11b to the front surface 11a of the wafer 11 are continuously formed along the first planned dividing lines 13a and the second planned dividing lines 13b. As a result, the wafer 11 is separated along the first planned dividing lines 13a and the second planned dividing lines 13b, and a plurality of device chips each having the devices 15 are obtained (see Fig. 1).

[0042] The cracks 25a formed in the first region 11c by the first modified layer forming step may not have reached the front surface 11a of the wafer 11. However, when cracks 25b are formed by the second modified layer forming step, reaching the back surface 11b of the wafer 11, the cracks 25a propagate to reach the front surface 11a of the wafer 11, using the formation of the cracks 25b as a trigger. Furthermore, the protective member 17 is already bonded to the front surface 11a side of the wafer 11 when the wafer 11 is separated. The layout of the individual device chips is thus maintained by the protective member 17 even after the wafer 11 is separated into the device chips.

[0043] Preferably, the number of first modified layers 23a formed in the first region 11c and the number of second modified layers 23b formed in the second region 11d are adjusted depending on the thickness, material, and / or the like of the wafer 11. For example, when the wafer 11 is a silicon wafer with a diameter of 12 inches and a thickness of 775 µm, it has been confirmed that the wafer 11 can be properly separated if the first modified layers 23a are formed up to twice in the first region 11c for each of the first planned separation lines 13a and the second planned separation lines 13b, and the second modified layers 23b are formed up to twice in the second region 11d for each of the first planned separation lines 13a and the second planned separation lines 13b.Furthermore, the number of first modified layers 23a formed in the first region 11c is preferably smaller than the number of modified layers 23b formed in the second region 11d. This numerical ratio between the first modified layers 23a and the second modified layers 23b has been confirmed to make the wafer 11 more resistant to the occurrence of warpage.

[0044] Next, the wafer 11 is ground on the back surface 11b side to thin the wafer 11 to a predetermined thickness (grinding step). In the grinding step, the wafer 11 is ground using, for example, a grinder. Fig. 5 is a side view illustrating a grinding device 12.

[0045] The grinding apparatus 12 includes a chuck table (holding table) 14 that holds the wafer 11. The chuck table 14 is connected to an unillustrated rotational drive source such as an electric motor, and this rotational drive source rotates the chuck table 14 about a rotational axis substantially parallel to the vertical direction. Furthermore, an unillustrated moving mechanism is disposed below the chuck table 14, and this moving mechanism moves the chuck table 14 in a processing feed direction. The chuck table 14 has an upper surface forming a holding surface 14a that holds the wafer 11. According to the shape of the wafer 11, the holding surface 14a is formed in a circular shape. However, the shape of the holding surface 14a can be appropriately changed depending on the shape of the wafer 11 or the like.The holding surface 14a is connected to a suction source (not shown) via a flow channel (not shown) formed inside the clamping table 14.

[0046] A grinding unit 16 is arranged above the chuck table 14. The grinding unit 16 includes an unillustrated spindle housing supported by an unillustrated lifting mechanism. A spindle 18 is housed in the spindle housing, and a disc-shaped holder 20 is fixed to a lower end portion of the spindle 18. On the lower surface side of the holder 20, a grinding wheel 22 having substantially the same diameter as the holder 20 is mounted. The grinding wheel 22 includes an annular receiver 24 formed with a metal material such as stainless steel or aluminum. Further, on the lower surface side of the receiver 24, a plurality of grindstones 26 formed in a cuboid shape are fixedly secured. The grindstones 26 are arranged in a circle along an outer periphery of the receiver 24.An unillustrated rotation drive source, such as an electric motor, is connected to an upper end side (a proximal end side) of the spindle 18. By a force transmitted from the rotation drive source, the grinding wheel 22 is rotated about a rotation axis substantially parallel to the vertical direction. Furthermore, an unillustrated grinding fluid supply passage is arranged inside the grinding unit 16 to supply a grinding fluid, such as pure water. The grinding fluid is supplied to the wafer 11 and the grindstones 26 while grinding is applied to the wafer 11.

[0047] In a grinding step, the wafer 11 is first placed on the chuck table 14 so that the front surface 11a side of the wafer 11 (the side of the protective member 17) and the holding surface 14a face each other. When a negative pressure of the suction source is applied to the holding surface 14a in this state, the wafer 11 is held by the chuck table 14 with the rear surface 11b side exposed upward via a negative pressure. Next, the chuck table 14 is moved and placed under the grinding unit 16. Then, both the chuck table 14 and the grinding wheels 22 are rotated, and the spindle 18 is caused to descend while supplying the grinding fluid to the rear surface 11b side of the wafer 11. The descending speed of the spindle 18 is adjusted so that the lower surfaces of the grinding stones 26 are pressed against the side of the rear surface 11b of the wafer 11 with an appropriate force.When the grinding stones 26 come into contact with the rear surface 11b side of the wafer 11, the wafer 11 is ground on the rear surface 11b side so that the wafer 11 is thinned. Then, grinding of the wafer 11 continues until the thickness of the wafer 11 is reduced to a predetermined thickness (target thickness). The target thickness corresponds to a finished thickness of the device chips obtained by dicing the wafer 11.

[0048] Fig. 6 is a sectional view showing a portion of the wafer 11 after the grinding step on an enlarged scale. When the grinding step is performed, the wafer 11, which has been separated along the planned separation lines 13a and the second planned separation lines 13b, is thinned to the target thickness.

[0049] When the first modified layers 23a (see Fig.3A, etc.) remain in the wafer 11 after grinding, the device chips obtained by dicing the wafer 11 have lower bending strength. Therefore, in the above-mentioned first modified layer forming step, it is preferable to form the lower first modified layers 23a so that the distance between the front surface 11a of the wafer 11 and the lower first modified layers 23a in the thickness direction of the wafer 11 becomes larger than the target thickness. In other words, the lower first modified layers 23a are not formed in a region whose distance (depth) from the front surface 11a of the wafer 11 is equal to or smaller than the target thickness.In the case of the preferred approach described above, all of the upper and lower first modified layers 23a are removed when the wafer 11 is ground during the grinding step until the thickness of the wafer 11 has been reduced to the target thickness. As a result, the first modified layers 23a no longer remain in the wafer 11 after grinding, thereby preventing a reduction in the bending strength of the device chips.

[0050] As described above, the wafer processing method according to the present embodiment includes the first modified layer forming step and the second modified layer forming step. In the first modified layer forming step, first modified layers 23a are formed along the first planned dividing lines 13a and the second planned dividing lines 13b. In the second modified layer forming step, second modified layers 23b are formed along the first planned dividing lines 13a and the second planned dividing lines 13b. In the wafer processing method described above, when forming the first modified layers 23a along the first planned dividing line 13a and the second planned dividing lines 13b in the second region 11d, no modified layers are yet formed.As a result, virtually no diffuse scattering (distribution) of the laser beam 8 occurs in the second region 11d, while the laser beam 8 is radiated onto the wafer 11 with the focal point of the laser beam 8 positioned in the first region 11c. As a result, there is hardly any radiation of the laser beam 8 onto the components 15, thus preventing damage to the components 15.

[0051] In the case of using a method that separates a wafer after forming modified layers in the wafer by using a stretchable film, or a method that separates a wafer by subjecting the wafer to a grinding process after forming modified layers in the wafer, as is common practice, it is possible that an external force may not be applied to the entire wafer as intended, and thus the wafer may not be properly separated. Furthermore, at the time a step of forming modified layers in a wafer has been performed, it is difficult to confirm whether the modified layers have been properly formed along planned separation lines, and thus it is difficult to predict whether the wafer will be properly separated in a subsequent step.On the other hand, in the wafer processing method according to the present embodiment, the wafer 11 is separated by utilizing cracks generated by forming modified layers from the front surface 11a side to the back surface 11b side of the wafer 11. As a result, the wafer 11 is properly separated along the first planned separation lines 13a and the second planned separation lines 13b. In addition, the separation step of the wafer 11 has already been completed by the time the first modified layer formation step and the second modified layer formation step have been performed. Therefore, before performing a subsequent step (grinding step or the like), it is possible to easily confirm whether the wafer 11 has been properly separated or not.

Claims

[1] A wafer processing method for processing a wafer (11) having components (15) formed in regions on one side of a front surface (11a) of the wafer (11), the regions being defined by first planned dividing lines (13a) and second planned dividing lines (13b) intersecting the first planned dividing lines (13a), the wafer processing method comprising: a protective element bonding step comprising bonding a protective element (17) to the front surface (11a) side of the wafer (11), a first modified layer forming step comprising irradiating a laser beam (8), the laser beam (8) having a transmissivity through the wafer (11), from a rear surface (11b) side of the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), with a focal point of the laser beam (8) set at a height that is at a height of a first region (11c) arranged inside the wafer (11), the wafer (11) being held via the protective member (17) by a chuck table (4) on the front surface (11a) side of the wafer (11), thereby forming first modified layers (23a) in the first region (11c), after performing the first modified layer forming step, a second modified layer forming step comprising irradiating the laser beam (8) from the rear surface (11b) side of the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), with the focal point of the laser beam (8) set at a height that lies at a height of a second region (11d) arranged inside the wafer (11) on the rear surface (11b) side of the wafer (11), whereby second modified layers (23b) are formed in the second region (11d) and cracks (25) are formed that extend from the front surface (11a) to the rear surface (11b) of the wafer (11) to separate the wafer (11) along the first planned dividing lines (13a) and the second planned dividing lines (13b), and after performing the second modified layer forming step, a grinding step comprising grinding the wafer (11) on the back surface (11b) side of the wafer (11) to thin the wafer (11) to a predetermined thickness. [2] A wafer processing method according to claim 1, wherein, during the first modified layer forming step, cracks are formed extending from the first modified layers (23a) to the front surface (11a) of the wafer (11). [3] Wafer processing method according to claim 1 or 2, wherein, during the first modified layer forming step, the first modified layers (23a) are formed such that the front surface (11a) of the wafer (11) and the first modified layers (23a) have a distance therebetween which is greater than the predetermined thickness, and during the grinding step, the wafer (11) is thinned to the predetermined thickness, thereby removing the first modified layers (23a).

Images