CN114670511A

CN114670511A - Double-sided metal-clad laminate and circuit board

Info

Publication number: CN114670511A
Application number: CN202111574183.4A
Authority: CN
Inventors: 田川和树; 矢熊建太郎; 安藤敏男
Original assignee: Nippon Steel and Sumikin Chemical Co Ltd
Current assignee: Nippon Steel Chemical and Materials Co Ltd
Priority date: 2020-12-24
Filing date: 2021-12-21
Publication date: 2022-06-28
Also published as: TW202224935A; JP2022101499A; KR20220092426A

Abstract

The invention provides a double-sided metal-clad laminated board and a circuit board, which have excellent dimensional stability and can also cope with large-scale size. A double-sided metal-clad laminate comprising an insulating resin layer and metal layers laminated on both sides of the insulating resin layer, wherein the length of the double-sided metal-clad laminate in the width direction perpendicular to the length direction is 230mm or more, and the double-sided metal-clad laminate is aligned with the insulating resin layer at a position in the length directionWhen at least three measurement sites of a plurality of measurement sites having different positions in the width direction are set to measure the birefringence in the thickness direction, the absolute value of the slope of a straight line obtained by approximating a plot point corresponding to each measurement site by the least square method is less than 1 × 10 in coordinates in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance in the width direction from an arbitrary reference position in the width direction to each measurement site is taken as the horizontal axis^‑5/mm。

Description

Double-sided metal-clad laminate and circuit board

Technical Field

The invention relates to a metal-clad laminate and a circuit board.

Background

In recent years, with the progress of downsizing, weight saving, and space saving of electronic devices, there has been an increasing demand for a Flexible Printed circuit board (FPC) that is thin and lightweight, has flexibility, and has excellent durability even when repeatedly bent. Since FPCs can be mounted in a limited space in a three-dimensional and high-density manner, their applications are expanding to components such as a Hard Disk Drive (HDD), a Digital Versatile Disk (DVD), a wiring or cable (cable) of a movable portion of an electronic device such as a mobile phone and a smartphone, and a connector (connector).

The FPC is manufactured by etching a metal layer of a metal-clad laminate having the metal layer and an insulating resin layer and performing wiring processing. In a photolithography (photolithography) process for mounting a metal-clad laminate or in a process of mounting an FPC, various processes such as bonding, cutting, exposure, etching, and the like are performed. The processing accuracy in these steps is important in maintaining the reliability of the electronic device mounted with the FPC.

However, since the metal-clad laminate has a structure in which metal layers having different coefficients of thermal Expansion (hereinafter sometimes referred to as "CTE") and insulating resin layers are laminated, internal stress is generated between the layers due to a difference in the Coefficients of Thermal Expansion (CTE) between the metal layers and the insulating resin layers. The internal stress is released when the metal layer is etched and wiring is performed, and the insulating resin layer expands and contracts, which causes a change in the size of the wiring pattern. In addition, in the process of manufacturing a double-sided metal-clad laminate by bonding a single-sided metal-clad laminate and a metal foil using a continuous pressing apparatus, strain is generated in the surface of the double-sided metal-clad laminate due to uneven pressurization at the time of thermocompression bonding by a pressing roller or misalignment between rollers arranged in a paper passing line from the roller to the winding, and dimensional stability is lowered.

As described above, if a dimensional change occurs at the stage of the circuit board, a connection failure between the wirings or between the wirings and the terminals is caused, which leads to a decrease in reliability and yield of the circuit board. Therefore, dimensional stability is a very important characteristic of a double-sided metal-clad laminate as a material of a circuit board.

As a technique for improving the dimensional stability of a metal-clad laminate, patent document 1 proposes that the thermal expansion coefficient of a polyimide insulating layer is within a predetermined range and that the product of the thickness of a rolled copper foil and the tensile elastic coefficient is within a predetermined range. In addition, patent document 2 proposes to reduce a deviation (Δ RO) between a value of the in-plane Retardation (RO) and the in-plane Retardation (RO) in a width Direction (Transverse Direction (TD)) Direction. Patent document 3 proposes a polarizing optical element and a polarization state control device that can adjust a phase difference generated between two linearly polarized lights whose polarization directions are orthogonal to each other.

[ Prior art documents ]

[ patent document ]

[ patent document 1] Japanese patent laid-open No. 2016-60138

[ patent document 2] Japanese patent laid-open publication No. 2017-200759

[ patent document 3] Japanese patent laid-open No. 2016-126804

Disclosure of Invention

[ problems to be solved by the invention ]

Since the demand for FPCs in electronic devices is expected to increase in the future, the demand for metal-clad laminates as materials thereof is increasing in order to increase the size, and particularly, to increase the length in the width direction, so that as many FPCs as possible can be processed and prepared. In this case, particularly in a double-sided metal-clad laminate produced by laminating a metal foil on a single-sided metal-clad laminate having a length in the width direction of 500mm or more, there is a problem that the dimensional change rate in the length direction of the insulating resin layer varies depending on the position in the width direction. Therefore, it is required not only to suppress the dimensional change rate in the longitudinal direction of the insulating resin layer in the double-sided metal-clad laminate to a small value, but also to minimize the fluctuation width so that the dimensional change rate is substantially constant even when the position in the width direction changes.

The present invention provides a double-sided metal-clad laminate which has a small variation width of the dimensional change rate in the longitudinal direction of an insulating resin layer regardless of the position in the width direction, has excellent dimensional stability, and can also cope with an increase in size.

[ means for solving the problems ]

As a result of diligent studies, the present inventors have found that the above problems can be solved by a double-sided metal clad laminate produced by laminating a metal foil on a single-sided metal clad laminate having a length in the width direction of 500mm or more using a continuous press apparatus and then controlling the tension balance in the width direction, and have completed the present invention.

That is, the double-sided metal-clad laminate according to the first aspect of the present invention is a long film-shaped double-sided metal-clad laminate including an insulating resin layer and metal layers laminated on both sides of the insulating resin layer, and

the length of the double-sided metal-clad laminate in the width direction orthogonal to the length direction is 230mm or more,

when the birefringence in the thickness direction is measured at least at three measurement sites set in the width direction of the insulating resin layer for a plurality of measurement sites having the same position in the length direction and different positions in the width direction in the insulating resin layer, in a coordinate in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance in the width direction from an arbitrary reference position in the width direction to each measurement site is taken as the horizontal axis, the absolute value of the slope of a straight line obtained by approximating the plotted point corresponding to each measurement site by the least square method Value less than 1 × 10^-5/mm。

The double-sided metal-clad laminate according to the second aspect of the present invention is a long film-shaped double-sided metal-clad laminate including an insulating resin layer and metal layers laminated on both sides of the insulating resin layer. In the double-sided metal-clad laminate according to the second aspect of the present invention, the length of the double-sided metal-clad laminate in the width direction orthogonal to the length direction is 500mm or more and 1200mm or less. In the double-sided metal-clad laminate according to the second aspect of the present invention, when measuring birefringence in the thickness direction at measurement portions that are respectively set at least three positions symmetrical with respect to a center line connecting midpoints of entire lengths in the width direction of the insulating resin layer at a plurality of measurement portions that are identical in position in the length direction and different in position in the width direction in the insulating resin layer, an absolute value of a slope of a straight line obtained by approximating plot points corresponding to the measurement portions by a least square method is less than 1 × 10 in coordinates where a value of birefringence in the thickness direction is set as a vertical axis and distances in the width direction from arbitrary reference positions in the width direction to the measurement portions are set as horizontal axes^-5/mm。

In the double-sided metal-clad laminate according to the first or second aspect of the present invention, the birefringence values in the thickness direction at all measurement sites may be 0.15 or less.

In the double-sided metal-clad laminate according to the first or second aspect of the present invention, when the insulating resin layer is divided into two virtual regions bounded by a center line connecting midpoints of the entire lengths in the width direction, all measurement positions may be set in a range from the center line to 49% of the entire length in the width direction in each of the two virtual regions.

In the double-sided metal-clad laminate of the first or second aspect of the present invention, the insulating resin layer may comprise a plurality of polyimide layers, and the metal layer is a copper layer.

The circuit board according to the present invention is obtained by processing one or both of the metal layers of the metal-clad laminates according to the first or second aspect into wiring.

[ Effect of the invention ]

The length of the double-sided metal-clad laminate in the width direction is 500mm or more when the double-sided metal-clad laminate is laminated, and the absolute value of the slope of an approximate straight line obtained by measuring the birefringence of the insulating resin layer in the thickness direction at different positions in the width direction is less than 1 x 10^-5And/mm. It is shown that the dimensional change in the longitudinal direction of the insulating resin layer is substantially stable regardless of the position in the width direction, and has excellent dimensional stability. Therefore, when the metal layer is processed into a circuit, the variation of the wiring interval caused in the surface of the double-sided metal-clad laminate, particularly in the processing portion in the width direction, can be suppressed to a minimum. Therefore, by using the double-sided metal-clad laminate of the present invention, the reliability of the circuit board can be improved while suppressing the reduction in yield.

Drawings

Fig. 1 is a diagram showing an external configuration of a double-sided metal-clad laminate according to an embodiment of the present invention.

Fig. 2 is an explanatory diagram of simulation coordinates showing a relationship between a value of birefringence in a thickness direction and a distance to a measurement site in a width direction.

Fig. 3 is an explanatory diagram showing the position of the measurement site on the horizontal axis of the coordinates shown in fig. 2.

Fig. 4 is a diagram for explaining a delay evaluation system used in the examples and comparative examples.

Fig. 5 is a schematic diagram for explaining a method of measuring the delay used in the examples and comparative examples.

Fig. 6 is a diagram for explaining a method of measuring a dimensional change rate after etching.

Fig. 7 is a graph showing the variation in the dimensional change rate in the width direction after etching of the double-sided metal-clad laminates obtained in example 1 and comparative example 1.

[ description of symbols ]

2: polyimide film

2 a: laminated surface

2 b: casting surface

10: measurement site

10 a: center of measurement site

20: test specimen

21: light source

22: light-receiving part

100: double-sided metal-clad laminated plate

100A: segment piece

A. B: virtual area

A. B, C, D, E, F: determining position

d: thickness/film thickness

θ₁: angle of incidence

θ₂: angle of refraction

Lo: center line

L₁、L₂、L₃: light (es)

N1-N10, S1-S10: hole marks

TD and MD: direction of rotation

Detailed Description

[ double-sided metal-clad laminate ]

Embodiments of the present invention will be described with reference to the accompanying drawings as appropriate. Fig. 1 shows an external appearance structure of a double-sided metal-clad laminate 100 according to an embodiment of the present invention. The double-sided metal-clad laminate 100 is in the form of a long film as a whole. Although not shown, the double-sided metal-clad laminate 100 includes an insulating resin layer and metal layers laminated on both sides of the insulating resin layer. In the following description, the longitudinal Direction of the long film-shaped double-sided metal-clad laminate 100 is sometimes referred to as the Machine Direction (MD) Direction, the width Direction orthogonal to the MD Direction is referred to as the TD Direction, and the axial Direction perpendicular to the plane (xy plane) formed by the MD Direction and the TD Direction is sometimes referred to as the thickness Direction (z Direction). The same applies to the insulating resin layer and the metal layer (metal foil) in the double-sided metal-clad laminate 100, the later-described segment 100A, and the sample 20.

The length of the double-sided metal-clad laminate 100 in the width direction (TD direction) perpendicular to the longitudinal direction may be 230mm or more, but in a preferred aspect thereof, the width (TD direction) may be in a range of 500mm or more and 1200mm or less. When the length in the TD direction is 500mm or more, the productivity in circuit processing of the double-sided metal-clad laminate 100 can be improved, but in general, it tends to be more difficult to control the dimensional stability and in-plane isotropy after lamination as the width in lamination is larger. That is, as the length in the TD direction during lamination increases, variation in the value of the birefringence Δ n (xy-z) measured in the thickness direction (hereinafter sometimes referred to as "Δ n (xy-z)") tends to occur in a plurality of portions in the same direction, and this aspect is improved and the variation in Δ n (xy-z) is small in the double-sided metal-clad laminate 100 of the present invention. Here, "birefringence Δ n (xy-z) in the thickness direction" means a difference between a refractive index Nxy in an in-plane direction (xy plane) and a refractive index Nz in a cross-sectional direction (z direction) orthogonal to the in-plane direction in the insulating resin layer. Since the orientation of molecules is more advanced, Δ n (xy-z) is increased because molecules tend to be aligned in the in-plane direction, and when no orientation is performed, Δ n (xy-z) is decreased. Therefore, Δ n (xy-z) can be used to evaluate the degree of orientation of the molecule. Further, the degree of molecular orientation affects the dimensional change in the MD direction after etching. That is, the degree of change in dimension in the MD direction after etching can be grasped from Δ n (xy-z). As described later, in the production method in which the single-sided metal-clad laminate and the metal foil are thermocompression bonded using the pressure roll in the roll-to-roll manner, a variation in Δ n (xy-z) tends to easily occur in the TD direction, and particularly, this tendency becomes remarkable when the length in the TD direction is 500mm or more. Accordingly, the effect of the present invention is particularly great in a double-sided metal-clad laminate having a width of 500mm or more at the time of lamination. If the width exceeds 1200mm, the dimensional stability and thickness variation in the surface becomes large, and for example, a defect is likely to occur when the film is processed into an FPC or the like, and the yield tends to be deteriorated.

< insulating resin layer >

The insulating resin layer preferably includes a non-thermoplastic polyimide layer and thermoplastic polyimide layers laminated on both surfaces of the non-thermoplastic polyimide layer. The insulating resin layer is preferably formed by a casting method in which a solution of thermoplastic polyimide or non-thermoplastic polyimide, or a solution of these precursors is sequentially applied. For example, when the insulating resin layer is formed by a casting method, a three-layer structure in which a thermoplastic polyimide layer, a non-thermoplastic polyimide layer, and a thermoplastic polyimide layer are sequentially stacked from the casting surface side of the insulating resin layer is preferable. The "casting surface" of the insulating resin layer refers to a surface on the metal layer side to which the polyimide or the precursor solution thereof is applied. In addition, a surface of the insulating resin layer opposite to the casting surface may be referred to as a "lamination surface".

In the present embodiment, as the non-thermoplastic polyimide constituting the non-thermoplastic polyimide layer and the thermoplastic polyimide constituting the thermoplastic polyimide layer, usual non-thermoplastic polyimide and thermoplastic polyimide can be used as the circuit board material without particular limitation. The "non-thermoplastic polyimide" herein is a polyimide which does not soften and does not exhibit adhesiveness even when heated in general, and in the present specification, it means a polyimide other than a thermoplastic polyimide, and preferably means that the storage elastic modulus at 30 ℃ measured with a dynamic viscoelasticity measuring apparatus (dynamic mechanical analyzer, DMA)) is 1.0 × 10 ⁹Pa or more and a storage modulus of elasticity at 280 ℃ of 3.0X 10⁸Polyimide having Pa or more. The term "thermoplastic polyimide" is usually a polyimide whose glass transition temperature (Tg) can be clearly confirmed, and in the present specification, it is preferable that the storage elastic modulus at 30 ℃ measured by DMA is 1.0X 10⁹Pa or more and a storage elastic modulus at 280 ℃ of less than 3.0X 10⁸Pa of a polyimide.

In the insulating resin layer, the thickness ratio ((a)/(B)) of the thickness (a) of the non-thermoplastic polyimide layer to the thickness (B) of the thermoplastic polyimide layer is preferably in the range of 1 to 20, and more preferably in the range of 2 to 12. When the number of non-thermoplastic polyimide layers and/or thermoplastic polyimide layers is plural, the thickness (a) or the thickness (B) refers to the total thickness. If the value of the ratio is less than 1, the non-thermoplastic polyimide layer becomes thin relative to the entire insulating resin layer, and therefore variation in the in-plane birefringence Δ n (x-y) (hereinafter sometimes referred to as "Δ n (x-y)") tends to increase, and if it exceeds 20, the thermoplastic polyimide layer becomes thin, and therefore the reliability of adhesion between the insulating resin layer and the metal layer tends to decrease. Here, Δ n (x-y) is a difference between two refractive indices Nx and Ny in the xy plane in the insulating resin layer. The refractive index Nxy in the in-plane direction (xy plane) is an average of the refractive index Nx in the x direction and the refractive index Ny in the y direction. Control of Δ n (x-y) is related to the resin structure of each polyimide layer constituting the insulating resin layer and the thickness thereof. The larger the thickness of the thermoplastic polyimide layer, which is a resin structure to which adhesiveness, i.e., high thermal expansion or softening, is given, the larger the influence on the value of Δ n (x-y) of the insulating resin layer. Therefore, it is preferable to increase the ratio of the thicknesses of the non-thermoplastic polyimide layers and decrease the ratio of the thicknesses of the thermoplastic polyimide layers, thereby reducing the value of Δ n (x-y) of the insulating resin layer from deviation thereof. As described later, in the present embodiment, even when the ratio of the thickness of the thermoplastic polyimide layer is reduced, the adhesion between the metal layer and the insulating resin layer can be ensured by designing the thermoplastic polyimide layer so as to contain a predetermined amount of diamine residue selected from the group consisting of the general formulae (2) and (3).

From the viewpoint of more effectively improving the dimensional accuracy of the insulating resin layer, the double-sided metal-clad laminate 100 of the present embodiment preferably has a width (length in the TD direction) of 500mm or more and a long length of 20m or more. As described later, after the double-sided metal-clad laminate 100 of the present embodiment is continuously produced, the long double-sided metal-clad laminate 100 may be cut at a constant value in the longitudinal direction (MD direction) and TD direction and used, and the slit-processed laminate may be included in the double-sided metal-clad laminate 100 of the present embodiment.

(non-thermoplastic polyimide)

In the present embodiment, the non-thermoplastic polyimide constituting the non-thermoplastic polyimide layer contains a tetracarboxylic acid residue and a diamine residue, and these preferably contain both aromatic groups, and more preferably, both the tetracarboxylic acid residue and the diamine residue contain only aromatic groups. Both the tetracarboxylic acid residue and the diamine residue contained in the non-thermoplastic polyimide contain an aromatic group, and therefore the ordered structure of the non-thermoplastic polyimide is easily formed, and variation in Δ n (x-y) can be suppressed while reducing the amount of change in Δ n (x-y) in the insulating resin layer under a high-temperature environment.

In the present invention, the tetracarboxylic acid residue means a tetravalent group derived from a tetracarboxylic dianhydride, and the diamine residue means a divalent group derived from a diamine compound. In addition, in the "diamine compound", the hydrogen atoms in the terminal two amino groups may be substituted, and may be, for example, -NR₃R₄(Here, R is₃、R₄Independently an optional substituent such as an alkyl group).

The tetracarboxylic acid residue contained in the non-thermoplastic polyimide is not particularly limited, and examples thereof include a tetracarboxylic acid residue derived from pyromellitic dianhydride (PMDA) (hereinafter also referred to as PMDA residue) and a tetracarboxylic acid residue derived from 3,3',4,4' -biphenyltetracarboxylic dianhydride (BPDA) (hereinafter also referred to as BPDA residue). These tetracarboxylic acid residues tend to form a rank structure, and the amount of change in Δ n (x-y) in a high-temperature environment can be reduced. The PMDA residue plays a role in controlling the thermal expansion coefficient and controlling the glass transition temperature. Further, since the BPDA residue has no polar group in the tetracarboxylic acid residue and has a large molecular weight, the effect of reducing the imide group concentration of the non-thermoplastic polyimide and suppressing moisture absorption of the insulating resin layer can be expected. From this viewpoint, the total amount of PMDA residues and/or BPDA residues may be preferably 50 parts by mole or more, more preferably in the range of 50 parts by mole to 100 parts by mole, and most preferably in the range of 70 parts by mole to 100 parts by mole, based on 100 parts by mole of all tetracarboxylic acid residues contained in the non-thermoplastic polyimide.

Examples of other tetracarboxylic acid residues contained in the non-thermoplastic polyimide include tetracarboxylic acid residues derived from the following aromatic tetracarboxylic dianhydrides: 2,3',3,4' -biphenyltetracarboxylic dianhydride, 2',3,3' -biphenyltetracarboxylic dianhydride, 3,3',4,4' -diphenylsulfonetetracarboxylic dianhydride, 4,4' -oxydiphthalic anhydride, 2',3,3' -benzophenonetetracarboxylic dianhydride, 2,3,3',4' -benzophenonetetracarboxylic dianhydride or 3,3',4,4' -benzophenonetetracarboxylic dianhydride, 2,3',3,4' -diphenylethertetracarboxylic dianhydride, bis (2, 3-dicarboxyphenyl) ether dianhydride, 3,3',4,4' -p-terphenyltetracarboxylic dianhydride, 2,3,3',4' -p-terphenyltetracarboxylic dianhydride or 2,2',3,3' -p-terphenyltetracarboxylic dianhydride, 2-bis (2, 3-dicarboxyphenyl) -propane dianhydride or 2, 2-bis (3, 4-dicarboxyphenyl) -propane dianhydride, bis (2, 3-dicarboxyphenyl) methane dianhydride or bis (3, 4-dicarboxyphenyl) methane dianhydride, bis (2, 3-dicarboxyphenyl) sulfone dianhydride or bis (3, 4-dicarboxyphenyl) sulfone dianhydride, 1-bis (2, 3-dicarboxyphenyl) ethane dianhydride or 1, 1-bis (3, 4-dicarboxyphenyl) ethane dianhydride, 1,2,7, 8-phenanthrene-tetracarboxylic acid dianhydride, 1,2,6, 7-phenanthrene-tetracarboxylic acid dianhydride or 1,2,9, 10-phenanthrene-tetracarboxylic acid dianhydride, 2,3,6, 7-anthracenetetracarboxylic acid dianhydride, 2-bis (3, 4-dicarboxyphenyl) tetrafluoropropane dianhydride, 2,3,5, 6-cyclohexane dianhydride, 1,2,5, 6-naphthalenetetracarboxylic dianhydride, 1,4,5, 8-naphthalenetetracarboxylic dianhydride, 2,3,6, 7-naphthalenetetracarboxylic dianhydride, 4, 8-dimethyl-1, 2,3,5,6, 7-hexahydronaphthalene-1, 2,5, 6-tetracarboxylic dianhydride, 2, 6-dichloronaphthalene-1, 4,5, 8-tetracarboxylic dianhydride or 2, 7-dichloronaphthalene-1, 4,5, 8-tetracarboxylic dianhydride, 2,3,6, 7-tetrachloronaphthalene-1, 4,5, 8-tetracarboxylic dianhydride or 1,4,5, 8-tetrachloronaphthalene-2, 3,6, 7-tetracarboxylic dianhydride, 2,3,8, 9-perylene-tetracarboxylic dianhydride, 3,4,9, 10-perylene-tetracarboxylic dianhydride, 4,5,10, 11-perylene-tetracarboxylic dianhydride or 5,6,11, 12-perylene-tetracarboxylic dianhydride, cyclopentane-1, 2,3, 4-tetracarboxylic dianhydride, pyrazine-2, 3,5, 6-tetracarboxylic dianhydride, pyrrolidine-2, 3,4, 5-tetracarboxylic dianhydride, thiophene-2, 3,4, 5-tetracarboxylic dianhydride, 4' -bis (2, 3-dicarboxyphenoxy) diphenylmethane dianhydride, and the like.

The diamine residue contained in the non-thermoplastic polyimide is preferably a diamine residue derived from a diamine compound represented by the following general formula (1) (hereinafter, sometimes referred to as "diamine residue (1)").

[ solution 1]

In the general formula (1), the linking group Z represents a single bond or-COO-, Y independently represents a C1-3 monovalent hydrocarbon group which may be substituted by halogen or phenyl, a C1-3 alkoxy group, or a C1-3 perfluoroalkyl group or alkenyl group, n represents an integer of 0-2, and p and q independently represent an integer of 0-4. The term "independently" as used herein means that the substituents Y, p and q in the formula (1) may be the same or different.

The diamine residue (1) is likely to form a ordered structure, and can improve dimensional stability, and particularly, can effectively suppress the amount of change in Δ n (x-y) in a high-temperature environment. From this viewpoint, the diamine residue (1) may be contained in a range of 20 parts by mole or more, preferably 70 parts by mole to 95 parts by mole, and more preferably 80 parts by mole to 90 parts by mole, based on 100 parts by mole of the total diamine residues contained in the non-thermoplastic polyimide.

Preferable specific examples of the diamine residue (1) include diamine residues derived from the following diamine compounds: p-phenylenediamine (p-phenylene diamine, p-PDA), 2'-dimethyl-4,4' -diaminobiphenyl (2,2'-dimethyl-4,4' -diaminobiphenyl, m-TB), 2'-diethyl-4,4' -diaminobiphenyl (2,2'-diethyl-4,4' -diaminobiphenyl, m-EB), 2'-diethoxy-4,4' -diaminobiphenyl (2,2'-diethoxy-4,4' -diaminobiphenyl, m-EOB), 2'-dipropoxy-4,4' -diaminobiphenyl (2,2 '-dipoxy-4, 4' -diaminobiphenyl, m-POB), 2'-n-propyl-4,4' -diaminobiphenyl (2,2' -n-propyl-4,4' -diaminobiphenol, m-NPB), 2' -divinyl-4,4' -diaminobiphenyl (2,2' -divinyl-4,4' -diaminobiphenol, VAB), 4' -diaminobiphenyl, 4' -diamino-2,2' -bis (trifluoromethyl) biphenyl (4,4' -diaminono-2, 2' -bis (trifluoromethylphenyl) biphenol, TFMB) and the like. Among these, 2'-dimethyl-4,4' -diaminobiphenyl (m-TB) is particularly preferable because it is easy to form a ordered structure and the amount of change in Δ n (x-y) in a high-temperature environment can be reduced.

In order to reduce the elastic coefficient of the insulating resin layer and improve the elongation, the bending resistance, and the like, it is preferable that the non-thermoplastic polyimide contains at least one diamine residue selected from the group consisting of diamine residues represented by the following general formulae (2) and (3).

[ solution 2]

In the above formulae (2) and (3), R₅、R₆、R₇And R₈Each independently represents a halogen atom, or an alkyl or alkoxy group or alkenyl group having 1 to 4 carbon atoms which may be substituted with a halogen atom, and X independently represents a group selected from-O-, -S-, -CH₂-、-CH(CH₃)-、-C(CH₃)₂-、-CO-、-COO-、-SO₂A divalent radical of-NH-or-NHCO-, X₁And X₂Each independently represents a bond, -O-, -S-, -CH₂-、-CH(CH₃)-、-C(CH₃)₂-、-CO-、-COO-、-SO₂A divalent radical of-NH-or-NHCO-, other than X₁And X₂In addition to the two are single bonds, m, n, o and p independently represent an integer of 0 to 4.

The term "independently" means that a plurality of linking groups X and X are present in one or both of the formulae (2) and (3)₁And X₂A plurality of substituents R₅、R₆、R₇、R₈And the integers m, n, o and p may be the same or different.

Since the diamine residues represented by the general formulae (2) and (3) have a flexible portion, flexibility can be imparted to the insulating resin layer. Here, since the number of benzene rings of the diamine residue represented by the general formula (3) is four, the terminal group bonded to the benzene ring is preferably a para-position in order to suppress an increase in the Coefficient of Thermal Expansion (CTE). In addition, from the viewpoint of imparting flexibility to the insulating resin layer and suppressing an increase in the Coefficient of Thermal Expansion (CTE), the diamine residues represented by the general formulae (2) and (3) may be contained in a range of preferably 5 to 30 parts by mole, more preferably 10 to 20 parts by mole, relative to 100 parts by mole of all the diamine residues contained in the non-thermoplastic polyimide. When the diamine residue represented by the general formulae (2) and (3) is less than 5 parts by mole, the insulating resin layer may have an increased elastic coefficient, a decreased elongation, a decreased bending resistance, or the like, and when it exceeds 30 parts by mole, the molecular orientation may be decreased, and it may be difficult to achieve a low CTE.

A diamine residue represented by the general formula (2)The group is preferably one or more of m, n and o being 0, and the group R being₅、R₆And R₇Preferable examples of the halogen atom-substituted alkyl group include an alkyl group having 1 to 4 carbon atoms which may be substituted with a halogen atom, an alkoxy group having 1 to 3 carbon atoms, and an alkenyl group having 2 to 3 carbon atoms. Further, in the general formula (2), preferable examples of the linking group X include-O-, -S-, -CH₂-、-CH(CH₃)-、-SO₂-or-CO-. Preferred specific examples of the diamine residue represented by the general formula (2) include diamine residues derived from the following diamine compounds: 1,3-bis (4-aminophenoxy) benzene (1,3-bis (4-aminophenoxy) bezene, TPE-R), 1,4-bis (4-aminophenoxy) benzene (1,4-bis (4-aminophenoxy) bezene, TPE-Q), bis (4-aminophenoxy) -2,5-di-tert-butylbenzene (bis (4-aminophenoxy) -2,5-di-tert-butylbenzene, DTBAB), 4-bis (4-aminophenoxy) benzophenone (4,4-bis (4-aminophenoxy) bezene, BAPK), 1,3-bis [2- (4-aminophenyl) -2-propyl-benzophenone]Benzene, 1,4-bis [2- (4-aminophenyl) -2-propyl]Benzene, and the like.

The diamine residue represented by the general formula (3) is preferably one or more of m, n, o and p in which 0 is contained, and the group R is preferably₅、R₆、R₇And R₈Preferred examples of the (C1-C4) alkyl group which may be substituted with a halogen atom, an alkoxy group having 1-C3 carbon atoms, or an alkenyl group having 2-C3 carbon atoms. In the general formula (3), the linking group X is ₁And X₂Preferred examples of (B) include a single bond, -O-, -S-, -CH₂-、-CH(CH₃)-、-SO₂-or-CO-. Wherein, from the viewpoint of imparting a bent portion, the linking group X is₁And X₂Except for the case where both are single bonds. Preferable specific examples of the diamine residue represented by the general formula (3) include diamine residues derived from the following diamine compounds: 4,4' -bis (4-aminophenoxy) biphenyl (4,4' -bis (4-aminophenoxy) biphenyl, BAPB), 2' -bis [4- (4-aminophenoxy) phenyl]Propane (2,2' -bis [4- (4-aminophenyloxy) phenyl ]]propane, BAPP), 2' -bis [4- (4-aminophenoxy) phenyl]Ether (2,2' -bis [4- (4-aminophenyloxy) phenyl)]ether, BAPE), bis [4- (4-aminophenoxy) phenyl]Sulfones, and the like.

Of the diamine residues represented by the general formula (2), particularly preferred is a diamine residue derived from 1, 3-bis (4-aminophenoxy) benzene (TPE-R) (sometimes referred to as "TPE-R residue"), and of the diamine residues represented by the general formula (3), particularly preferred is a diamine residue derived from 2,2' -bis [4- (4-aminophenoxy) phenyl ] propane (BAPP) (sometimes referred to as "BAPP residue"). Since TPE-R residues and BAPP residues have a flexible portion, the elastic coefficient of the insulating resin layer can be reduced and flexibility can be imparted. Further, since BAPP has a large molecular weight, an effect of reducing the imide group concentration of the non-thermoplastic polyimide and suppressing moisture absorption of the insulating resin layer can be expected.

Examples of the other diamine residue contained in the non-thermoplastic polyimide include diamine residues derived from the following aromatic diamine compounds: m-phenylenediamine (m-PDA), 4,4' -diaminodiphenyl ether (4,4' -diaminodiphenyl ether, 4,4' -DAPE), 3' -diaminodiphenyl ether, 3,4' -diaminodiphenyl ether, 4,4' -diaminodiphenylmethane, 3' -diaminodiphenylmethane, 3,4' -diaminodiphenylmethane, 4,4' -diaminodiphenylpropane, 3' -diaminodiphenylpropane, 3,4' -diaminodiphenylpropane, 4,4' -diaminodiphenylsulfide, 3' -diaminodiphenylsulfide, 3,4' -diaminodiphenylsulfide, 4,4' -diaminodiphenylsulfone, 3' -diaminodiphenylsulfone, 3,4' -diaminodiphenylsulfone, 4,4 '-diaminobenzophenone, 3' -diaminobenzophenone, 2-bis- [4- (3-aminophenoxy) phenyl ] propane, bis [4- (3-aminophenoxy) phenyl ] sulfone, bis [4- (3-aminophenoxy) biphenyl, bis [1- (3-aminophenoxy) ] biphenyl, bis [4- (3-aminophenoxy) phenyl ] methane, bis [4- (3-aminophenoxy) phenyl ] ether, bis [4- (3-aminophenoxy) ] benzophenone, 9-bis [4- (3-aminophenoxy) phenyl ] fluorene, 2-bis- [4- (4-aminophenoxy) phenyl ] hexafluoropropane, 2, 2-bis- [4- (3-aminophenoxy) phenyl ] hexafluoropropane, 3' -dimethyl-4, 4' -diaminobiphenyl, 4' -methylenedi-o-toluidine, 4' -methylenedi-2, 6-dimethylaniline, 4' -methylene-2, 6-diethylaniline, 3' -diaminodiphenylethane, 3' -diaminobiphenyl, 3' -dimethoxybenzidine, 3 "-diamino-p-terphenyl, 4' - [1, 4' -phenylenebis (1-methylethylidene) ] dianiline, 4' - [1, 3-phenylenebis (1-methylethylidene) ] dianiline, a salt thereof, and a salt thereof, Bis (p-aminocyclohexyl) methane, bis (p-beta-amino-t-butylphenyl) ether, bis (p-beta-methyl-delta-aminopentyl) benzene, p-bis (2-methyl-4-aminopentyl) benzene, p-bis (1, 1-dimethyl-5-aminopentyl) benzene, 1, 5-diaminonaphthalene, 2, 6-diaminonaphthalene, 2, 4-bis (beta-amino-t-butyl) toluene, 2, 4-diaminotoluene, m-xylene-2, 5-diamine, p-xylene-2, 5-diamine, m-xylylenediamine, p-xylylenediamine, 2, 6-diaminopyridine, 2, 5-diamino-1, 3, 4-oxadiazole, and the like, Piperazine, and the like.

In the non-thermoplastic polyimide, the thermal expansion coefficient, storage elastic coefficient, tensile elastic coefficient, and the like can be controlled by selecting the kinds of the tetracarboxylic acid residue and the diamine residue, or the molar ratio of each of the tetracarboxylic acid residue and the diamine residue when two or more kinds of the tetracarboxylic acid residue and the diamine residue are applied. In the case of a non-thermoplastic polyimide having a plurality of polyimide structural units, the polyimide structural units may be present in the form of blocks or may be present randomly, but are preferably present randomly from the viewpoint of suppressing variations in Δ n (x-y).

The imide group concentration of the non-thermoplastic polyimide is preferably 35% by weight or less. Here, "imide group concentration" means the imide group (- (CO) in polyimide₂A value obtained by dividing the molecular weight of-N-) by the molecular weight of the entire structure of the polyimide. When the imide group concentration exceeds 35% by weight, the molecular weight of the resin itself becomes small and the low hygroscopicity is also deteriorated due to the increase of the polar group. By selecting the combination of the acid anhydride and the diamine compound, the molecular orientation of the non-thermoplastic polyimide is controlled, and thereby the increase in CTE associated with the decrease in the imide group concentration is suppressed, and low hygroscopicity is ensured.

(thermoplastic polyimide)

In the present embodiment, the thermoplastic polyimide constituting the thermoplastic polyimide layer contains a tetracarboxylic acid residue and a diamine residue, and these preferably contain both aromatic groups, and more preferably contain only aromatic groups in all of the tetracarboxylic acid residue and the diamine residue. Both of the tetracarboxylic acid residue and the diamine residue contained in the thermoplastic polyimide contain an aromatic group, and thus the amount of change in Δ n (x-y) in the insulating resin layer in a high-temperature environment can be suppressed.

The tetracarboxylic acid residue contained in the thermoplastic polyimide is not particularly limited, and examples thereof include a tetracarboxylic acid residue derived from pyromellitic dianhydride (PMDA) (hereinafter also referred to as PMDA residue) and a tetracarboxylic acid residue derived from 3,3',4,4' -biphenyltetracarboxylic dianhydride (BPDA) (hereinafter also referred to as BPDA residue). These tetracarboxylic acid residues easily form a ordered structure, and the amount of change in Δ n (x-y) in a high-temperature environment can be reduced. The PMDA residue plays a role in controlling the thermal expansion coefficient and controlling the glass transition temperature. Further, since the BPDA residue has no polar group in the tetracarboxylic acid residue and has a large molecular weight, the effect of reducing the imide group concentration of the thermoplastic polyimide and suppressing moisture absorption of the insulating resin layer can be expected. From this viewpoint, the total amount of PMDA residues and/or BPDA residues may be preferably 50 parts by mole or more, more preferably in the range of 50 parts by mole to 100 parts by mole, and most preferably in the range of 70 parts by mole to 100 parts by mole, based on 100 parts by mole of all tetracarboxylic acid residues contained in the thermoplastic polyimide.

Examples of the other tetracarboxylic acid residue contained in the thermoplastic polyimide include tetracarboxylic acid residues derived from the same aromatic tetracarboxylic dianhydrides as those exemplified in the non-thermoplastic polyimide.

In the present embodiment, the diamine residue contained in the thermoplastic polyimide is preferably at least one diamine residue selected from the group consisting of the above-mentioned general formulae (2) and (3). The diamine residue selected from the group consisting of the diamine residues represented by the general formulae (2) and (3) is preferably 50 parts by mole or more, more preferably 50 parts by mole to 100 parts by mole, and most preferably 70 parts by mole to 100 parts by mole, based on 100 parts by mole of the total diamine residues. By including 50 parts by mole or more of the diamine residues selected from the group consisting of the general formulae (2) and (3) in total per 100 parts by mole of all the diamine residues, flexibility and adhesiveness can be imparted to the thermoplastic polyimide layer, and the thermoplastic polyimide layer can function as an adhesive layer to the metal layer. Among the diamine residues represented by the general formula (2), TPE-R residues are particularly preferable, and BAPP residues are particularly preferable among the diamine residues represented by the general formula (3). Since TPE-R residues and BAPP residues have flexible portions, the elastic coefficient of the insulating resin layer can be reduced, and flexibility can be imparted. Further, since BAPP has a large molecular weight, an effect of reducing the imide group concentration of thermoplastic polyimide and suppressing moisture absorption of the insulating resin layer can be expected.

In addition, as described above, in the case where the non-thermoplastic polyimide constituting the non-thermoplastic polyimide layer contains a diamine residue selected from the group consisting of the general formula (2) and the general formula (3), the thermoplastic polyimide constituting the thermoplastic polyimide layer may contain a similar structure, preferably the same kind of diamine residue selected from the group consisting of the general formula (2) and the general formula (3) as the diamine residue. In this case, although the ratio of the diamine residue contained in the thermoplastic polyimide and the non-thermoplastic polyimide is different, when the similar or the same kind of diamine residue is contained, particularly when a polyimide film is formed by a casting method, the alignment control of the thermoplastic polyimide layer and the non-thermoplastic polyimide layer becomes easy, and the dimensional accuracy can be easily controlled. From this viewpoint, in the present embodiment, it is preferable that both the non-thermoplastic polyimide constituting the non-thermoplastic polyimide layer and the thermoplastic polyimide constituting the thermoplastic polyimide layer contain at least one diamine residue selected from the group consisting of the above-mentioned general formulae (2) and (3), and most preferably contain a TPE-R residue and/or a BAPP residue.

In the present embodiment, examples of the diamine residue other than the diamine residues represented by the general formulae (2) and (3) contained in the thermoplastic polyimide include diamine residues derived from the following diamine compounds: 2,2' -dimethyl-4, 4' -diaminobiphenyl (m-TB), 2' -diethyl-4, 4' -diaminobiphenyl (m-EB), 2' -diethoxy-4, 4' -diaminobiphenyl (m-EOB), 2' -dipropoxy-4, 4' -diaminobiphenyl (m-POB), 2' -di-n-propyl-4, 4' -diaminobiphenyl (m-NPB), 2' -divinyl-4, 4' -diaminobiphenyl (VAB), 4' -diaminobiphenyl, 4' -diamino-2, 2' -bis (trifluoromethyl) biphenyl (TFMB), p-phenylenediamine (p-PDA), m-phenylenediamine (m-PDA), and, 3,3' -diaminodiphenylmethane, 3' -diaminodiphenylpropane, 3' -diaminodiphenylsulfide, 3' -diaminodiphenylsulfone, 3-diaminodiphenylether, 3,4' -diaminodiphenylmethane, 3,4' -diaminodiphenylpropane, 3,4' -diaminodiphenylsulfide, 3' -diaminobenzophenone, (3,3' -diamino) diphenylamine, 1, 4-bis (3-aminophenoxy) benzene, 3- [4- (4-aminophenoxy) phenoxy ] aniline, 3- [3- (4-aminophenoxy) phenoxy ] aniline, 1,3-bis (3-aminophenoxy) benzene (1,3-bis (3-aminophenoxy) bezene, APB), 4'- [ 2-methyl- (1, 3-phenylene) dioxy ] dianiline, 4' - [ 4-methyl- (1, 3-phenylene) dioxy ] dianiline, 4'- [ 5-methyl- (1, 3-phenylene) dioxy ] dianiline, bis [4- (3-aminophenoxy) phenyl ] methane, bis [4- (3-aminophenoxy) phenyl ] propane, bis [4- (3-aminophenoxy) phenyl ] ether, bis [4- (3-aminophenoxy) phenyl ] sulfone, bis [4- (3-aminophenoxy) ] benzophenone, bis [4,4' - (3-aminophenoxy) ] benzanilide, phenol, aniline, and aniline, and aniline, wherein the compound [4-, 4- [3- [4- (4-aminophenoxy) phenoxy ] aniline, 4' - [ oxybis (3, 1-phenyleneoxy) ] dianiline, bis [4- (4-aminophenoxy) phenyl ] ether (BAPE), bis [4- (4-aminophenoxy) phenyl ] ketone (BAPK), bis [4- (3-aminophenoxy) ] biphenyl, bis [4- (4-aminophenoxy) ] biphenyl and the like.

In the thermoplastic polyimide, the thermal expansion coefficient, the tensile elastic coefficient, the glass transition temperature, and the like can be controlled by selecting the kinds of the tetracarboxylic acid residue and the diamine residue, or by selecting the molar ratio of each of the tetracarboxylic acid residue and the diamine residue when two or more kinds of the tetracarboxylic acid residue and the diamine residue are applied. In the case where the thermoplastic polyimide has a plurality of polyimide structural units, the thermoplastic polyimide may be present in the form of blocks or may be present randomly, but is preferably present randomly.

The thermoplastic polyimide constituting the thermoplastic polyimide layer can improve adhesion to the metal layer. The glass transition temperature of the thermoplastic polyimide is in the range of 200 ℃ to 350 ℃, preferably 200 ℃ to 320 ℃.

The imide group concentration of the thermoplastic polyimide is preferably 35% by weight or less. Here, the "imide group concentration" means an imide group (- (CO) in polyimide₂The molecular weight of the (E-N-) is divided by the molecular weight of the polyimide as a wholeThe value of (c). When the imide group concentration exceeds 35% by weight, the molecular weight of the resin itself becomes small and the low hygroscopicity is also deteriorated due to the increase of the polar group. By controlling the molecular orientation of the thermoplastic polyimide by selecting the combination of the acid anhydride and the diamine compound, an increase in CTE associated with a decrease in the imide group concentration is suppressed, and low hygroscopicity is ensured.

(Synthesis of non-thermoplastic polyimide and thermoplastic polyimide)

Generally, polyimide can be manufactured by: tetracarboxylic dianhydride and a diamine compound are reacted in a solvent to produce a polyamic acid, and then heated to be closed in a ring. For example, a tetracarboxylic dianhydride and a diamine compound are dissolved in an organic solvent in approximately equimolar amounts, and the resulting solution is stirred at a temperature in the range of 0 to 100 ℃ for 30 minutes to 24 hours to cause a polymerization reaction, thereby obtaining a polyamic acid as a precursor of a polyimide. In the reaction, the reaction components are dissolved in the organic solvent so that the amount of the precursor formed is in the range of 5 to 30 wt%, preferably 10 to 20 wt%. Examples of the organic solvent used in the polymerization reaction include: n, N-Dimethylformamide (DMF), N-dimethylacetamide (DMAc), N-diethylacetamide, N-methyl-2-pyrrolidone (NMP), 2-butanone, Dimethylsulfoxide (DMSO), hexamethylphosphoramide, N-methylcaprolactam, Dimethyl sulfate, cyclohexanone, dioxane, tetrahydrofuran, diglyme (diglyme), triglyme, cresol, and the like. Two or more of these solvents may be used in combination, and an aromatic hydrocarbon such as xylene or toluene may be used in combination. The amount of the organic solvent used is not particularly limited, but is preferably adjusted so that the concentration of the polyamic acid solution obtained by the polymerization reaction is about 5 to 30 wt%.

The polyamic acid synthesized is generally advantageously used as a reaction solvent solution, but may be concentrated, diluted or replaced with another organic solvent as necessary. In addition, polyamic acid is generally excellent in solvent solubility and therefore can be used favorably. The solution of the polyamic acid preferably has a viscosity in the range of 500cps to 100,000 cps. If the amount is outside the above range, defects such as uneven thickness and streaks are likely to occur in the film during coating work using a coater or the like. The method for imidizing the polyamic acid is not particularly limited, and for example, heat treatment such as heating at a temperature in the range of 80 to 400 ℃ for 1 to 24 hours is suitably employed.

The weight average molecular weight of the polyimide is, for example, preferably within a range of 10,000 to 400,000, and more preferably within a range of 50,000 to 350,000. When the weight average molecular weight is less than 10,000, the strength of the insulating resin layer tends to be reduced and the insulating resin layer tends to be brittle. On the other hand, when the weight average molecular weight exceeds 400,000, the viscosity tends to increase excessively, and defects such as uneven thickness and streaks tend to occur during coating operation.

The CTE of the insulating resin layer as a whole is preferably 1X 10 ^-630 x 10,/K or more^-6In the range of/K or less. If the CTE is less than 1X 10^-6Or more than 30X 10^-6When the metal-clad laminates 100 are used in the present embodiment, the metal-clad laminates are not easily deformed in the TD direction. The CTE of the non-thermoplastic polyimide layer may preferably be 1X 10^-6/K～30×10^-6In the range of/K, the CTE of the thermoplastic polyimide layer may preferably exceed 30X 10^-6And is at 80X 10^-6In the range of/K or less. The polyimide layer can be formed to have a desired thermal expansion coefficient by appropriately changing the combination of raw materials used, the thickness, the drying and the curing conditions.

The thickness of the insulating resin layer may be set to a thickness within a predetermined range according to the thickness, rigidity, or the like of the metal layer, and is, for example, preferably within a range of 6 to 50 μm, and more preferably within a range of 9 to 38 μm. If the thickness of the insulating resin layer is less than the lower limit, problems may occur such as failure to secure electrical insulation properties, or difficulty in handling in the manufacturing process due to a decrease in workability. On the other hand, if the thickness of the insulating resin layer exceeds the above upper limit, for example, the bending resistance when the FPC is bent may be reduced.

When the width (length in the TD direction) of the insulating resin layer is in the range of 500mm to 1200mm, when Δ n (xy-z) is measured at a plurality of measurement sites set at the same position in the MD direction and at different positions in the TD direction, that is, at least three measurement sites are set at positions symmetrical with respect to a center line connecting midpoints of the entire length of the insulating resin layer in the TD direction, the absolute value of the slope of a straight line obtained by approximating plot points corresponding to the measurement sites by the least square method is less than 1 × 10 in coordinates where the value of Δ n (xy-z) is taken as the vertical axis and the distance in the TD direction from an arbitrary reference position in the TD direction to each measurement site is taken as the horizontal axis ^-5And/mm. This point will be described below with reference to fig. 2.

Fig. 2 shows a simulation of coordinates when Δ n (xy-z) of the insulating resin layer is measured at a plurality of measurement positions set at different positions in the TD with respect to the double-sided metal-clad laminate 100 of the present embodiment having a width (length in the TD direction) in a range of 500mm to 1200 mm. In fig. 2, the ordinate represents the value Δ n (xy-z), and the abscissa represents the measurement position in the TD direction with the one end of the double-sided metal-clad laminate 100 as a reference position. Here, the results of measuring positions at 6 are shown at A, B, C, D, E, F. The horizontal axis may indicate the relative distance in the TD direction from an arbitrary reference position to each measurement position. Therefore, the "reference position" may be any position in the TD direction of the insulating resin layer, and may be, for example, one end in the TD direction, the opposite end thereof, a midpoint thereof, or the like. The measurement position is not limited to 6 positions in total, and is preferably 6 to 10 positions, for example.

The upper straight line in fig. 2 is a straight line obtained by approximating the set of plotted points of a circle corresponding to each measurement position of the double-sided metal-clad laminate 100 of the present embodiment by the least square method, and the lower straight line is a straight line obtained by approximating the set of plotted points of four directions corresponding to each measurement position of the double-sided metal-clad laminate to be compared by the least square method. Both the upper and lower straight lines can be expressed by the formula y ═ ax + b, where a denotes the slope of the straight line, and b denotes an arbitrary value of Δ n (xy-z). Since fig. 2 is not an actual measurement result but a graph of a simulation for explanation, the value of Δ n (xy-z), the slope of a straight line, and the like are not strict.

In fig. 2, the absolute value | a | of the slope of the upper line is smaller than 1 × 10^-5And/mm. In this way, by setting the slope of the approximate straight line obtained from the values of Δ n (xy-z) at the plurality of measurement sites in the TD direction within a predetermined range, it is possible to suppress the variation in the dimension change rate in the MD direction in the TD direction, and to improve the dimensional stability after etching. That is, since the variation in the value of Δ n (xy-z) in the TD direction of the double-sided metal-clad laminate 100 corresponding to the upper straight line is very small, it indicates that the dimensional change in the MD direction after etching is substantially constant regardless of the position in the TD direction, and the variation width of the dimensional change is very small.

On the other hand, in fig. 2, the absolute value | a | of the slope of the lower line obtained by approximating the square plotted points by the least square method is 1 × 10^-5More than mm. Therefore, the variation in Δ n (xy-z) values at a plurality of measurement sites in the TD direction of the double-sided metal-clad laminate to be compared is large, and the dimensional change in the MD direction after etching differs depending on the position in the TD direction, and the fluctuation range is large.

The double-sided metal-clad laminate 100 and the comparative double-sided metal-clad laminate were manufactured by a casting method. In fig. 2, the values of Δ n (xy-z) themselves are smaller in the square plotted point group than in the circle plotted point group regardless of the measurement positions a to F, and therefore it is considered that the both-side metal-clad laminate to be compared is excellent in dimensional stability in the MD direction after etching as a whole. However, as understood from the slopes of the two approximate straight lines, the square plotted point group has a larger deviation Δ n (xy-z) in the TD direction than the circular plotted point group, and therefore, there is a high possibility that the wiring width during circuit processing differs depending on the position in the TD direction. In this way, if the wiring widths of the manufactured FPCs are different depending on the in-plane processing positions of the double-sided metal-clad laminate, the uniformity of quality among the plurality of FPCs is lost, which causes a problem in terms of reliability. On the other hand, in the case of the set of plotted points of the circle corresponding to the double-sided metal-clad laminate 100 of the present embodiment, the value of Δ n (xy-z) is substantially constant regardless of the position in the TD direction, and the variation is very small, so that even if a dimensional change in the MD direction occurs after etching, for example, it is possible to achieve a countermeasure such as uniformly providing a margin at the time of circuit design in anticipation of the dimensional change. Therefore, the double-sided metal-clad laminate 100 has an advantage that it is easy to stabilize the quality (particularly, uniformity of the wiring width) between a plurality of FPCs to be processed, as compared with a double-sided metal-clad laminate to be compared.

In addition, regarding the insulating resin layer, when Δ n (xy-z) is measured for a plurality of measurement sites set at the same position in the MD direction and at different positions in the TD direction, it is preferable that Δ n (xy-z) has a value of 0.15 or less for all measurement sites. All the measurement sites Δ n (xy-z) showed a low value of 0.15 or less, indicating that dimensional change in the MD direction after etching of the insulating resin layer was small and dimensional stability was high. Such excellent dimensional stability is difficult to achieve in a metal-clad laminate using a polyimide film produced by a tenter method. In fig. 2, for the purpose of explanation, four sets of plot points are given as the objects of comparison, but since the metal-clad double-sided laminate of the objects of comparison is manufactured by a casting method, the dimensional accuracy in circuit processing is extremely superior to that of a metal-clad double-sided laminate using a polyimide film manufactured by a tenter method.

As described above, in the double-sided metal-clad laminate 100 of the present embodiment, although the width (length in the TD direction) is in the range of 500mm or more and 1200mm or less, Δ n (xy-z) shows a low value of 0.15 or less at all measurement sites, and therefore, not only is the dimensional stability in the MD direction after etching extremely high, but also the variation in Δ n (xy-z) in the TD direction is suppressed as much as possible, and therefore, the variation in the dimensional change in the MD direction after etching is extremely small in the TD direction. Therefore, the quality of the FPCs processed from the double-sided metal-clad laminate 100 can be stabilized, and the reliability of the FPCs can be improved.

On the abscissa of the graph of fig. 2, measurement positions at 6 points A, B, C, D, E, F are shown at equal intervals. The measurement position at 6 is the position of the measurement site in the TD direction of the double-sided metal-clad laminate 100. Specifically, as shown in fig. 3, from a segment 100A obtained by cutting the double-sided metal-clad laminated sheet 100 to have an arbitrary length in the MD direction, a rectangular region of 6 points is set as the measurement site 10 with the position in the MD direction being aligned. The position of the center 10a of each measurement site 10 (sample 20) in the rectangular shape corresponds to the "measurement position" on the horizontal axis in fig. 2. Rectangular measurement sites 10 were cut out from the evaluation film from which the metal layer was removed from the segment 100A by etching, and the resultant was used as a test piece (sample 20 described later; see fig. 4) for measuring Δ n (xy-z). In fig. 3, the dimensions of the respective portions are exaggerated for the sake of explanation.

In fig. 3, if there are 6 measurement sites 10 that are deviated in the TD direction, it is difficult to accurately grasp the deviation of Δ n (xy-z) in the TD direction, and therefore, it is preferable that the intervals between the measurement positions (centers 10a) of the respective measurement sites 10 be substantially equal.

In order to distribute the measurement sites 10 over a wide range in the TD, the distance between the measurement positions (centers 10a) of adjacent measurement sites 10 is preferably at least 1/12 or more, more preferably 1/10 or more, and most preferably 1/8 or more, over the entire length in the TD. The upper limit of the distance between adjacent measurement positions (centers 10a) may be set to at least six measurement positions in total in the TD direction, but considering that the distance between the measurement positions (centers 10a) is not equal, the upper limit may be set to 1/5 or less, and more preferably 1/6 or less, over the entire TD direction.

Further, as shown in fig. 3, when the insulating resin layer of the double-sided metal-clad laminate 100 is divided into two virtual regions A, B each bounded by a center line Lo connecting midpoints of the entire length in the TD direction, it is preferable to select the positions of the measurement regions 10 such that the measurement regions 10 present in the respective virtual regions A, B are located at positions symmetrical with respect to the center line Lo. In fig. 3, the measurement sites 10 at 6 are set at 3 positions respectively at symmetrical positions in each of the virtual areas A, B with the center line Lo as a reference, but 4 or more positions may be set at symmetrical positions in each of the virtual areas a and B.

Therefore, it is most preferable that the measurement positions (centers 10a) of the respective measurement regions 10 are spaced at substantially equal intervals, the interval between adjacent measurement positions (centers 10a) is 1/12 or more over the entire length in the TD direction, and the measurement regions 10 existing at 3 or more positions in the respective virtual regions A, B are located at positions symmetrical with respect to the center line Lo.

As shown in fig. 3, it is preferable that the measurement positions (center 10a) of all the measurement sites 10 are set in a range from the center line Lo to both ends of each of the two virtual regions A, B to 49% of the total length in the TD direction. As described later, in the manufacturing method in which the single-sided metal-clad laminate and the metal foil are thermally pressure-bonded using the pressing roll in the roll-to-roll manner, there is a possibility that the variation in Δ n (xy-z) in the vicinity of both end portions in the TD direction of the insulating resin layer in the double-sided metal-clad laminate 100 becomes large, and the vicinity of both end portions is cut off and not used in circuit processing.

Therefore, the center 10a of all the measurement sites 10 is set to the total length in the TD direction from the center line Lo to both ends preferably in the range of 0% to 49%, more preferably in the range of 0% to 45%, and most preferably in the range of 0% to 40%. In other words, the centers 10a of all the measurement sites 10 can be set within a range of preferably 98% excluding both end portions in the vicinity of each end portion, more preferably 90% excluding 5% excluding each end portion, and most preferably 80% excluding 10% excluding each end portion, from the entire length in the TD direction.

Further, since it is not preferable that the plurality of measurement sites 10 are offset in the TD direction, the distance between the centers 10a of two measurement sites 10 located outermost among the plurality of measurement sites 10 may be set to be in a range of preferably 50% or more, more preferably 60% or more, and still more preferably 70% or more, with respect to the total length in the TD direction.

From the above, in order to accurately grasp the variation in Δ n (xy-z) in the TD direction, the plurality of measurement sites 10 may be distributed such that the distance between the centers 10a of the two outermost measurement sites 10 among the plurality of measurement sites 10 is preferably in the range of 50% to 98% with respect to the entire length in the TD direction, more preferably in the range of 60% to 98%, and still more preferably in the range of 70% to 98%.

As shown in the examples described below, the test piece (sample 20) may be cut out from a segment obtained by dividing the segment 100A into two parts along the center line Lo. Alternatively, the test piece (sample 20) may be further cut out from small pieces obtained by cutting out both end portions of the piece 100A in the TD direction in a range of 1% to 20%. In this case, either end of the small segment in the TD direction may be set as the reference position.

< Metal layer >

Examples of the metal constituting the metal layer include metals selected from copper, aluminum, stainless steel, iron, silver, palladium, nickel, chromium, molybdenum, tungsten, zirconium, gold, cobalt, titanium, tantalum, zinc, lead, tin, silicon, bismuth, indium, and alloys thereof. Copper foil is particularly preferable in terms of conductivity. The copper foil may be any of an electrolytic copper foil and a rolled copper foil. When the double-sided metal-clad laminate 100 of the present embodiment is continuously produced, a long metal foil formed by winding a metal foil having a predetermined thickness into a roll shape is used as the metal foil. The length of the metal foil in the TD direction that can be applied to the double-sided metal-clad laminate 100 is preferably in the range of 500mm to 1200mm, for example. The thickness of the metal foil is, for example, preferably in the range of 6 to 20 μm, and more preferably in the range of 8 to 13 μm. In the double-sided metal-clad laminate 100, the two metal layers may have the same or different structures.

[ method for producing double-sided copper-clad laminate ]

A method for manufacturing the double-sided metal-clad laminate 100 will be described with reference to a double-sided copper-clad laminate in which a metal layer is a copper layer (copper foil) as an example. In this case, the copper layer laminated on one surface of the insulating resin layer of the double-sided copper-clad laminate is referred to as a "first copper layer", and the copper layer laminated on the opposite surface is referred to as a "second copper layer".

(step of laminating an insulating resin layer on the first copper layer)

For example, a laminate of an insulating resin layer and a first copper foil, that is, a single-sided copper-clad laminate having a width (length in the TD direction) in the range of 500mm to 1200mm, is formed by a so-called casting method in which a polyamic acid solution is directly applied to a first copper foil having a width (length in the TD direction) in the range of 500mm to 1200mm, and then dried and cured by heat treatment to form an insulating resin layer. The first copper foil in the single-sided copper-clad laminate is used as a first copper layer. In the case of forming an insulating resin layer including a plurality of polyimide layers by a casting method, a precursor layer may be formed by sequentially applying a coating liquid of polyamic acid and drying. For example, when the polyimide layer has a 3-layer structure, a method of applying a coating solution of polyamic acid in order to sequentially laminate a precursor layer of thermoplastic polyimide, a precursor layer of non-thermoplastic polyimide, and a precursor layer of thermoplastic polyimide on the first copper foil, drying the coating solution, and then performing heat treatment to imidize the coating solution is preferable. The first copper foil is not particularly limited, and a rolled copper foil or an electrolytic copper foil which is commercially available can be used.

(Process for laminating second copper layer)

The single-sided copper-clad laminate having the first copper layer and the separately prepared long second copper foil are passed through a press apparatus having two pairs of press rolls so as to face each other in a roll-to-roll conveying process. The unwinding tension of the single-sided copper-clad laminate is preferably in the range of 10N to 35N, and more preferably in the range of 15N to 30N. The unwinding tension of the second copper foil is preferably in the range of 5N to 25N, and more preferably in the range of 10N to 20N. The line speed of paper passing is preferably 2m/min to 10 m/min. In this way, by laminating the second copper foil on the insulating resin layer side of the single-sided copper-clad laminate by hot pressing with a press roll, a double-sided copper-clad laminate having a width (length in the TD direction) in the range of 500mm to 1200mm and having the first copper layer on one side of the insulating resin layer and the second copper layer on the other side can be obtained. The double-sided copper-clad laminate obtained by pressing is conveyed and wound on a rotating roll by a winding tension in a roll-to-roll manner, the balance of the winding tension applied in the TD direction of the double-sided copper-clad laminate can be controlled by adjusting the angle or height of the rotating roll, and the orientation of the insulating resin layer can be controlled by adjusting the stress applied to the double-sided copper-clad laminate.

The second copper foil used for the second copper layer is not particularly limited, and may be, for example, a rolled copper foil or an electrolytic copper foil. As the second copper foil, the same copper foil as the first copper foil may be used.

Although not shown in the drawings, the present embodiment may include a laminate obtained by cutting the double-sided metal-clad laminate 100 having a width (length in the TD direction) in the range of 500mm to 1200mm in the longitudinal direction and setting the length in the width direction orthogonal to the longitudinal direction to, for example, 230mm or more, preferably 230mm to 450mm, more preferably 230mm to 270mm, and most preferably 230mm to 250mm, as a modification. The divided metal-clad laminate of the modification can be produced by performing the following steps: the double-sided metal-clad laminate 100 having a width in the range of 500mm to 1200mm, which is obtained by the manufacturing method including the step of laminating the insulating resin layer on the first copper layer and the step of laminating the second copper layer, is further cut in the longitudinal direction.

In the divided double-sided metal-clad laminate, when measuring the birefringence in the thickness direction at a plurality of measurement sites having the same position in the longitudinal direction and different positions in the width direction in the insulating resin layer, that is, at least three measurement sites set in the width direction of the insulating resin layer, the absolute value of the slope of a straight line obtained by approximating a plot point corresponding to each measurement site by the least square method is less than 1 × 10 in coordinates in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance from an arbitrary reference position in the width direction to each measurement site in the width direction is taken as the horizontal axis ^-5The thickness is only mm.

Although not shown in the drawings, in the divided double-sided metal-clad laminate, if at least three measurement sites are deviated in the TD direction, it is difficult to accurately grasp the deviation of Δ n (xy-z) in the TD direction, and therefore, it is preferable that the intervals of the measurement positions of the respective measurement sites are substantially equal.

In order to distribute the measurement sites over a wide range in the TD, the distance between the measurement positions of adjacent measurement sites is preferably at least 1/6 or more, more preferably 1/5 or more, and most preferably 1/4 or more, over the entire length in the TD. The upper limit of the interval between adjacent measurement positions may be set to at least three measurement sites in total in the TD direction, but considering that the interval between the measurement positions is not equal, the upper limit may be preferably 2/5 or less, and more preferably 1/3 or less, with respect to the total length in the TD direction.

Further, in the divided double-sided metal-clad laminated sheet, the measurement positions of all the measurement sites may be set within a range of preferably 49% of the total length in the TD direction from one end portion to the other end portion, more preferably within a range of 0% to 45%, and most preferably within a range of 0% to 40%.

A preferable example of the divided double-sided metal clad laminate is a two-divided double-sided metal clad laminate obtained by cutting a double-sided metal clad laminate 100 having a width of about 500mm in the longitudinal direction along a center line Lo (see fig. 3) connecting midpoints of the entire lengths in the TD direction, and making the length in the width direction orthogonal to the longitudinal direction, for example, 230mm or more, preferably 230mm or more and 250mm or less. In the above case, when aligning the two-part double-sided metal-clad laminates in such a manner that the cut portions are in contact with each other and aligning the positions in the longitudinal direction with the same positions as those before cutting, when measuring the birefringence in the thickness direction at a plurality of measurement portions in the insulating resin layer, which are the same in the longitudinal direction and different in the width direction, that is, at least three measurement portions each set at a position symmetrical with respect to the cut portion, the absolute value of the slope of a straight line obtained by approximating the plotted points corresponding to the respective measurement portions by the least square method is preferably less than 1 × 10 in coordinates in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance from an arbitrary reference position in the width direction to the respective measurement portions is taken as the horizontal axis ^-5/mm。

The divided double-sided metal-clad laminate of the modification as described above is the same as the double-sided metal-clad laminate 100 except that the length in the width direction is different, and therefore is included in the double-sided metal-clad laminate 100 of the present embodiment unless otherwise noted.

[ Circuit Board ]

The double-sided metal-clad laminate 100 is effectively used mainly as a material for a circuit board such as an FPC. For example, a circuit board such as an FPC, a multilayer circuit board obtained by stacking the circuit boards in multiple layers, a rigid flexible board (rigid FPC), or the like, which is one embodiment of the present invention, can be manufactured by patterning the metal layer of the double-sided metal-clad laminate 100 to form a wiring layer by a conventional method.

[ examples ]

The following examples are provided to more specifically explain the features of the present invention. The scope of the present invention is not limited to the examples. In the following examples, unless otherwise specified, various measurements and evaluations were carried out by the following methods.

[ measurement of viscosity ]

The viscosity at 25 ℃ was measured using an E-type viscometer (product name: DV-II + Pro, manufactured by Brookfield corporation). The rotational speed was set so that the torque (torque) became 10% to 90%, and after 2 minutes from the start of measurement, the value at which the viscosity became stable was read.

[ measurement of weight average molecular weight ]

The weight average molecular weight was measured by gel permeation chromatography (manufactured by Tosoh (TOSOH) Co., Ltd., trade name: HLC-8220 GPC). Polystyrene was used as a standard and N, N-dimethylacetamide was used as a developing solvent.

[ measurement of glass transition temperature (Tg) ]

The glass transition temperature was measured from 30 ℃ to 400 ℃ with a temperature rise rate of 4 ℃/min and a frequency of 11Hz on a polyimide film having a size of 5 mm. times.20 mm using a dynamic viscoelasticity measuring apparatus (DMA: manufactured by UBM Co., Ltd., trade name: E4000F), and the temperature at which the elastic modulus (tan. delta.) is the maximum was defined as the glass transition temperature.

[ measurement of Coefficient of Thermal Expansion (CTE) ]

A polyimide film having a size of 3mm × 20mm was heated from 30 ℃ to 265 ℃ at a constant heating rate while applying a load of 5.0g thereto, and further held at the temperature for 10 minutes, and then cooled at a rate of 5 ℃/minute, using a thermomechanical analyzer (product name: 4000SA, manufactured by Bruker Co., Ltd.), to thereby determine an average thermal expansion coefficient (thermal expansion coefficient) of 250 ℃ to 100 ℃.

[ measurement of storage modulus of elasticity ]

A polyimide film having a size of 5 mm. times.20 mm was measured at a temperature rise rate of 4 ℃/min and a frequency of 11Hz from 30 ℃ to 400 ℃ using a dynamic viscoelasticity measuring apparatus (DMA: manufactured by UBM Co., Ltd., trade name: E4000F).

[ calculation of Birefringence (. DELTA.n (xy-z)) in thickness direction ]

The birefringence Δ n (xy-z) in the thickness direction was measured using a birefringence meter (product name: wide range birefringence evaluation system WPA-100, manufactured by Photonic-Lattice Co., Ltd.), measuring region: length direction (MD): 20mm × width direction (TD): 15 mm). The retardation Re described later is measured by a known polarization state control device (see, for example, patent document 3), and the birefringence Δ n (xy-z) in the thickness direction is calculated from the measurement result.

First, a method of evaluating the delayed Re will be described. Fig. 4 is an explanatory view showing a part of an evaluation system of the delay Re, and fig. 5 is a schematic view of a method of measuring the delay Re.

The system for evaluating retardation Re includes a birefringence/phase difference evaluation device (WPA-100, manufactured by Photonic-Lattice) and a device for changing the incident angle θ of light incident on a sample₁And a not-shown rotation device for rotating the sample. In fig. 4, reference numeral 20 denotes a sample, reference numeral 21 denotes a light source of the birefringence/phase difference evaluation device, and reference numeral 22 denotes a light receiving portion of the birefringence/phase difference evaluation device. The wavelength of light emitted from the light source 21 is 543 nm. The sample 20 is fixed to a not-shown rotating device in a state of being supported by a fixing frame.

The delay Re is obtained by using a delay not shownThe rotating device changes the inclination angle of the sample 20 supported by the frame to make the incident angle theta of the light incident on the sample 20₁The measurement was performed while changing (see fig. 5). Make the incident angle theta₁The changes were 0 °, ± 30 °, ± 40 °, ± 50 °, and the retardation Re was measured at each angle.

Next, a method for calculating the birefringence Δ n (xy-z) in the thickness direction will be described. The birefringence Δ n (xy-z) in the thickness direction was calculated using the measurement result of the retardation Re. An incident angle θ when the polyimide film is evaluated using the retardation evaluation system₁Angle of refraction theta₂As shown in fig. 5. In fig. 5, reference numeral 2 denotes a polyimide film including an insulating resin layer of a metal-clad laminate, 2a denotes a lamination surface of the

polyimide film

2, 2b denotes a casting surface of the polyimide film 2, and d denotes a thickness of the polyimide film. Here, with the notation L₁Indicating light before incidence on the lamination surface 2a, by a symbol L₂Light in the polyimide film 2 is shown by symbol L₃Light emitted from the casting surface 2b is shown. The X, Y and Z axes are orthogonal to each other, the XY direction is an axis parallel to the lamination surface 2a of the polyimide film 2, and the Z direction is an axis orthogonal to the lamination surface 2a of the polyimide film 2 and is an axis in the thickness direction.

As shown in the following formula (A), the retardation Re depends on the thickness d, the birefringence [ Delta ] n (xy-z) in the thickness direction, and the refraction angle [ theta ]₂. Angle of refraction theta₂Dependent on the angle of incidence theta₁. Thus, according to the angle of incidence θ for a plurality of angles of incidence₁The birefringence Δ n (xy-z) can be calculated from the measured values of the retardation Re.

Re＝d·Δn(xy-z)·sin²θ₂/cosθ₂…(A)

Wherein the angle of refraction theta₂The angle formed by the light beam inside the polyimide film 2 and the film normal line, the incident angle theta₁According to Snell's law, becomes theta₂＝sin^-1(sinθ₁The relationship of/N). Here, d is the film thickness, and N is the refractive index of the sample to be measured.

Further, Δ n (xy-z) is a difference between a refractive index in an in-plane direction and a refractive index in a thickness direction, and

Δ n (xy-z) ═ Nxy-Nz is satisfied.

Nxy: refractive index in-plane direction

Nz: refractive index in thickness direction

[ preparation of evaluation sample ]

Using 2 evaluation samples cut from the double-sided metal-clad laminate to a length direction (MD) of 200mm × a width direction (TD) of 250mm, a measurement site of birefringence in the thickness direction by the method was set at a position 6 in the width direction. The size of each measurement site was a rectangle 20mm in length × 15mm in width, and the distance from one end of each of the 2 parallel evaluation samples to the center of each measurement site was set as the "measurement position".

Specifically, when the length of the double-sided metal-clad laminate in the width direction was 540mm, the end portions were cut into 20mm pieces each to have a width of 500mm, the central portion was cut into 2 small pieces of the evaluation sample having a width of 250mm in the length direction along the center line Lo (see fig. 3), and the metal layer of each small piece was removed by etching to prepare 2 evaluation films. 2 films for evaluation were arranged so that the cut portions were in contact with each other, and the longitudinal positions were aligned with the same positions as those before cutting, and a rectangle having a length of 20mm × a width of 15mm was cut from one of the film ends so as to be the measurement positions shown in tables 1, 3, 5, 7, and 9 described later, thereby preparing a sample 20 for birefringence in the thickness direction. At this time, 3 samples 20, each of which was 6 in total, were prepared from positions symmetrical in the TD direction with respect to the boundary (corresponding to the original center line Lo) of 2 evaluation films. The measurement position was defined as the center of the sample 20 (center position 20mm in length × 15mm in width), and the birefringence Δ n (xy-z) in the thickness direction was measured.

[ measurement of dimensional Change Rate ]

In the double-sided copper-clad laminate, 2 evaluation samples were obtained by cutting the laminate to a length direction (MD) of 200mm × a width direction (TD) of 250 mm. Using this sample, as shown in FIG. 6, holes of φ 1mm were drilled at intervals of 100mm in the MD direction at 2 points and at intervals of φ 1mm in the TD direction at 50mm at 10 points using an NC drill, resulting in a total of 20 points. Using the sample, the center coordinate position of each circle was measured using a non-contact Computer Numerical Control (CNC) image measuring machine (manufactured by Mitutoyo corporation, trade name: Quick Vision QV-X404 PIL-C). Specifically, coordinates of 360 points are obtained at an angle of 1 degree on the circumference of each hole of 1mm, and the center coordinates of the circle are calculated by the least square method using the 360 point data.

The copper foils on both sides of the double-sided copper-clad laminate were removed by chemical etching, thereby forming a polyimide film. After the polyimide film was humidified at a temperature of 23 ℃ and a humidity of 50% RH for 20 hours or more, the center coordinate position of each circle was measured again by the same apparatus. From the obtained coordinate positions, the distance between adjacent holes in the MD direction (about 100mm) was calculated, and the dimensional change rate [% ] in the MD direction after etching based on the calculation formula before etching was calculated.

Calculating formula:

[ (distance after etching-distance before etching) ÷ distance before etching ]. times.100

When the length of the width of the double-sided copper-clad laminate is 500mm, the evaluation is performed by dividing the laminate into two parts in the width direction, and the dimensional change rate in the MD direction is evaluated by extracting the maximum value and the minimum value from all data in the width direction, and using the difference as an index of the variation (fluctuation width) of the dimensional change rate in the TD direction.

The abbreviations used in the examples and comparative examples represent the following compounds.

m-TB: 2,2 '-dimethyl-4, 4' -diaminobiphenyl

TPE-R: 1, 3-bis (4-aminophenoxy) benzene

BAPP: 2, 2-bis [4- (4-aminophenoxy) phenyl ] propane

And (3) PMDA: pyromellitic dianhydride

BPDA: 3,3',4,4' -biphenyltetracarboxylic dianhydride

And (3) DAPE: 4,4' -diaminodiphenyl ether

DMAc: n, N-dimethyl acetamide

(Synthesis example 1)

23.0 parts by weight of m-TB (0.108 parts by mole), 3.5 parts by weight of TPE-R (0.012 parts by mole), and DMAc (15% by weight solid content concentration after polymerization) were charged into a reaction vessel under a nitrogen gas stream, and dissolved by stirring at room temperature. Subsequently, 26.0 parts by weight of PMDA (0.119 parts by mole) was added thereto, and then the mixture was stirred at room temperature for 3 hours to carry out a polymerization reaction, thereby obtaining a polyamic acid solution a. The solution viscosity of polyamic acid solution a was 41,100 cps. A polyimide film having a thickness of 25 μm formed from the polyamic acid solution a (Tg: 421 ℃ C., CTE: 10ppm/K) was non-thermoplastic.

(Synthesis example 2)

30.2 parts by weight of BAPP (0.074 parts by mole) and DMAc in an amount such that the solid content concentration after polymerization became 15% by weight were put into a reaction vessel under a nitrogen gas flow, and the mixture was stirred and dissolved at room temperature. Subsequently, 22.3 parts by weight of BPDA (0.076 parts by mole) was added, and then the mixture was stirred at room temperature for 3 hours to perform a polymerization reaction, thereby obtaining a polyamic acid solution b. The solution viscosity of the polyamic acid solution b was 9,800 cps. The polyimide film having a thickness of 25 μm formed from the polyamic acid solution b (Tg: 252 ℃, CTE: 46ppm/K) was thermoplastic.

[ example 1]

The polyamic acid solution b prepared in Synthesis example 2 was uniformly applied to a copper foil 1 (rolled copper foil, long strip, thickness: 12 μm, length in the width direction: 540mm) so that the cured thickness became 2.5 μm, and then heated and dried at 120 ℃ to remove the solvent. The polyamic acid solution a prepared in synthesis example 1 was uniformly applied thereon so that the cured thickness became 20 μm, and then heated and dried at 120 ℃ to remove the solvent. Further, polyamic acid solution b prepared in synthesis example 2 was uniformly applied thereon so that the cured thickness became 2.5 μm, and then dried by heating at 120 ℃ to remove the solvent. Then, a stepwise heat treatment is performed at 130 ℃ to 360 ℃ to complete imidization, thereby producing a single-sided copper-clad laminate 1. A copper foil 1 is disposed on the surface of the polyimide layer in the single-sided copper-clad laminate 1, and the single-sided copper-clad laminate 1 is unwound under the following tension: 30N, roll surface temperature: 300-400 ℃ and linear pressure of a pressing roller: a transport speed (line speed) in the range of 38.6kgf/cm to 115.8 kgf/cm: thermocompression bonding was continuously performed under a condition of 4.0 m/min. With respect to the laminated board after thermocompression bonding, the tension balance in the width direction by the rotating roll was controlled with the winding tension 130N, thereby preparing the double-sided copper-clad laminated board 1.

Table 1 shows the measured positions and calculated values of the birefringence in the thickness direction of the double-sided copper-clad laminate 1.

[ Table 1]

As shown in Table 1, the slope of the approximate straight line calculated from the obtained 6-point Δ n (xy-z) values was-4X 10^-7/mm。

Next, the measured values of the hole distances before and after etching and the dimensional change rates of the hole marks N1 to N10 and S1 to S10 (see fig. 6) in the double-sided copper-clad laminate 1 are shown in table 2. The expression in the column of "hole mark" in table 2 refers to the distance between N1 and S1 (the distance between the centers of the holes) in fig. 6, for example, if it is "N1-S1". Table 4, table 6, table 8, table 10 are the same.

The variation in the TD-direction dimensional change rate after etching based on the reference value before etching in the double-sided copper-clad laminate 1 was 0.019.

[ example 2]

A double-sided copper-clad laminate 2 was produced in the same manner as in example 1, except that the copper foil used on both sides was a copper foil 2 (rolled copper foil, long, 18 μm in thickness, 540mm in length in the width direction).

Table 3 shows the measured positions and calculated values of the birefringence in the thickness direction of the double-sided copper-clad laminate 2.

[ Table 3]

As shown in Table 3, the slope of the approximate straight line calculated from the value of Δ n (xy-z) at 6 points was-2X 10 ^-6/mm。

Next, the measured values of the hole distances before and after etching and the dimensional change rates of the hole marks N1 to N10 and S1 to S10 in the double-sided copper-clad laminate 2 are shown in table 4.

[ Table 4]

The variation in the TD-direction dimensional change rate of the double-sided copper-clad laminate 2 in the MD direction after etching based on the pre-etching is 0.011.

[ example 3]

A double-sided copper-clad laminate 3 was produced in the same manner as in example 1, except that the copper foil 2 was used as the copper foil used on both sides, the thickness of the cured polyamic acid solution b was 2.5 μm, and the thickness of the cured polyamic acid solution a was 7 μm.

Table 5 shows the measured positions and calculated values of the birefringence in the thickness direction of the double-sided copper-clad laminate 3.

[ Table 5]

As shown in Table 5, the slope of the approximate straight line calculated from the value of Δ n (xy-z) at 6 points was-7X 10^-6/mm。

Next, the measured values of the hole distances before and after etching and the dimensional change ratios of the hole marks N1 to N10 and S1 to S10 in the double-sided copper-clad laminate 3 are shown in table 6.

[ Table 6]

The dimensional change rate in the MD direction after etching based on the value before etching in the double-sided copper-clad laminate 3 was 0.027 in the TD direction.

[ example 4]

A double-sided copper-clad laminate 4 was produced in the same manner as in example 1, except that copper foil 3 (electrolytic copper foil, long, 12 μm in thickness, 540mm in length in the width direction) was used as the copper foil used on both sides, the thickness of the cured polyamic acid solution b was 2 μm, and the thickness of the cured polyamic acid solution a was 46 μm.

Table 7 shows the measured positions and calculated values of the birefringence in the thickness direction of the double-sided copper-clad laminate 4.

[ Table 7]

As shown in Table 7, the slope of the approximate straight line calculated from the value of Δ n (xy-z) at 6 points was-2X 10^-7/mm。

Next, the measured values of the hole distances before and after etching and the dimensional change ratios of the hole marks N1 to N10 and S1 to S10 in the double-sided copper-clad laminate 4 are shown in table 8.

[ Table 8]

The variation in the TD-direction dimensional change rate of the double-sided copper-clad laminate 4 in the MD direction after etching based on the value before etching was 0.006.

Comparative example 1

A double-sided copper-clad laminate 5 was produced in the same manner as in example 1, except that the balance of the tension in the width direction generated by the rotating roll was not controlled in the laminate after thermocompression bonding.

Table 9 shows the measured positions and calculated values of the birefringence in the thickness direction of the double-sided copper-clad laminate 5.

[ Table 9]

As shown in Table 9, the slope of the approximate straight line calculated from the value of Δ n (xy-z) at 6 points was 1 × 10^-5/mm。

Next, the measured values of the hole distances before and after etching and the dimensional change ratios of the hole marks N1 to N10 and S1 to S10 in the double-sided copper-clad laminate 5 are shown in table 10.

[ Table 10]

The variation in the TD-direction dimensional change rate of the double-sided copper-clad laminate 5 in the MD direction after etching based on the value before etching was 0.062.

Fig. 7 shows a graph (vertical axis: dimensional change rate [% ], horizontal axis: TD direction position [ mm ]) comparing the variation in the TD direction of the MD direction dimensional change rate of example 1 (double-sided copper-clad laminate 1) and comparative example 1 (double-sided copper-clad laminate 5). The numbers on the horizontal axis in fig. 7 correspond to the numbers of the hole marks N1 to N10 and S1 to S10 in fig. 6.

The embodiments of the present invention have been described above in detail for the purpose of illustration, but the present invention is not limited to the embodiments and can be variously modified.

Claims

1. A double-sided metal-clad laminate comprising an insulating resin layer and metal layers laminated on both sides of the insulating resin layer, wherein the metal layers are in the form of a long film,

when the birefringence in the thickness direction is measured at least at three measurement sites set in the width direction of the insulating resin layer for a plurality of measurement sites having the same position in the length direction and different positions in the width direction in the insulating resin layer, the absolute value of the slope of a straight line obtained by approximating a plotted point corresponding to each measurement site by the least square method is less than 1 × 10 in coordinates in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance in the width direction from an arbitrary reference position in the width direction to each measurement site is taken as the horizontal axis ^-5/mm。

2. A double-sided metal-clad laminate which is a long film-shaped double-sided metal-clad laminate comprising an insulating resin layer and metal layers laminated on both sides of the insulating resin layer,

the length of the double-sided metal-clad laminate in the width direction orthogonal to the longitudinal direction is 500mm to 1200mm,

when measuring the birefringence in the thickness direction at a plurality of measurement sites having the same position in the longitudinal direction and different positions in the width direction in the insulating resin layer, wherein at least three measurement sites are set at symmetrical positions with respect to a center line connecting midpoints of the entire lengths in the width direction of the insulating resin layer, the absolute value of the slope of a straight line obtained by approximating a plot point corresponding to each measurement site by the least square method is less than 1 × 10 in coordinates in which the value of the birefringence in the thickness direction is taken as the vertical axis and the distance in the width direction from an arbitrary reference position in the width direction to each measurement site is taken as the horizontal axis^-5/mm。

3. The two-sided metal-clad laminate according to claim 1 or 2, wherein the birefringence index in the thickness direction of all measurement portions has a value of 0.15 or less.

4. The double-sided metal-clad laminate according to claim 1 or 2, wherein when the insulating resin layer is divided into two virtual regions bounded by a center line connecting midpoints of full lengths in the width direction, all measurement positions are set in a range from the center line to 49% of the full length in the width direction in each of the two virtual regions.

5. The two-sided metal-clad laminate according to claim 1 or 2, wherein the insulating resin layer comprises a plurality of polyimide layers, and the metal layer is a copper layer.

6. A circuit board obtained by processing one or both of the metal layers in the double-sided metal-clad laminate according to claim 1 or 2 into wiring.