JP3920048B2 - Through-hole filling paste - Google Patents

Description

【００１７】
一般式（１）におけるＲ₁としては、水素原子、炭素数１〜１０のアルキル基、炭素数１〜１０のヒドロキシアルキル基、または、炭素数１〜１０のアルキルオキシ基が好ましい置換基として挙げられる。更には、水素原子、炭素数１〜４のアルキル基（メチル基、エチル基、プロピル基、イソプロピル基、ブチル基）、炭素数１〜４のヒドロキシアルキル基（ヒドロキシメチル基、ヒドロキシエチル基、ヒドロキシプロピル基、ヒドロキシブチル基）、または、炭素数１〜４のアルキルオキシ基（メトキシ基、エトキシ基、プロポキシ基、イソプロポキシ基、ブトキシ基）が好ましい置換基として挙げられる。
【００１８】
特に好ましいイミダゾール系硬化剤は、Ｒ₁がメチル基である「２−フェニル−４−メチル−５−ヒドロキシメチル−イミダゾール」である。この硬化剤は、銅粉末を添加したスルーホール充填材で発生しやすい上記の隙間の問題を効果的に抑制できる。また、スルーホール内に充填したスルーホール充填用ペーストをキュアする際に発生する、スルーホール充填材の表裏面の凹みの問題を効果的に抑制することができる。スルーホール充填材の表裏面に凹みが発生すると、次いで形成されるビルドアップ層に凹凸を発生させる原因となるため好ましくない。特に、図７に示すようなＦＣ−ＰＧＡ型の配線基板のようにスルーホール直下のビルドアップ層にピン付けする場合にも問題となる。ここにいう「ＦＣ−ＰＧＡ型の配線基板」とは、半導体素子をＦＣ（フリップチップ）実装するとともに、マザーボードへの実装をピンによって行うタイプのＰＧＡ（ピングリッドアレイ）型配線基板をいう。
【００１９】
特には、アスペクト比が大きいスルーホール（１．６〜３２、好ましくは２．５〜３２、より好ましくは５〜３２、特には、８〜３２）や、スルーホール導体の内壁面にエッチングや黒化処理により十点平均粗さ（ＪＩＳ Ｂ ０６０１の３．５．１に準ずる。）による粗度Ｒzが０．３〜３μｍ（好ましくは２〜３μｍ）の凹凸を有する粗化面が形成されていたり、更に内壁面にうねりや段差があって直線性に乏しいようなスルーホール（レーザにより穿設されたもの等）や、直径の異なるスルーホールが混在（例えば、１００μｍと３００μｍが混在）する場合の各スルーホールを充填する場合において有用である。尚、スルーホールの内壁面には導通用のスルーホール導体が形成されているのがよいが、導電性を有するスルーホール充填用ペーストを用いる場合は、スルーホール導体の形成を省略することもできる。
【００２０】
本発明に用いるスルーホールを有する基板としては、ＦＲ−４、ＦＲ−５、ＢＴ等のいわゆるコア基板を用いるのがよいが、ＰＴＦＥ等の熱可塑性樹脂シートに厚み３５μm程度の厚手の銅箔を挟み込んでコア基板としたものにスルーホールを形成したものを用いてもよい。また、コア基板の少なくとも一面に絶縁層を形成した基板を貫通するように形成されたスルーホールを有する基板を用いることができる。図１１は、コア基板の両面に絶縁層を形成した基板を貫通するように形成されたスルーホールを有する基板の例である。この場合スルーホールは、絶縁層を形成した後に炭酸ガスレーザ等を用いたレーザ加工機を用いて形成するのがよい。しかし、レーザにより穿設されたスルーホールは、内壁面にうねりや段差があって直線性に乏しいため、スルーホール充填が困難な場合が多い。本発明で用いるスルーホール充填用ペーストは、コア基板の両面に絶縁層を形成した基板を貫通するようにレーザにより穿設されたスルーホールであっても、良好な充填を行うことができる利点がある。
【００２１】
本発明のプリント配線板の製造方法により得られたプリント配線板上に、絶縁層及び導体層を交互に積層したビルドアップ層を形成して多層プリント配線板を得ることができる。この場合、図７に示すようなコンデンサ内蔵型の多層プリント配線板であっても、いわゆるガラス−エポキシ複合材料（絶縁基板）の厚みを４００μｍ程度と、通常品の８００μｍの半分にまで薄くして低背化を図ることができる利点がある。また、スルーホール充填材の内部に隙間が無く、かつ、スルーホール近傍のビルドアップ層の平坦性に優れた基板を用いるため、図７に示すようなＦＣ−ＰＧＡ型の配線基板のようにスルーホール直下のビルドアップ層にピン付けしたＰＧＡ（ピングリッドアレイ）型の多層プリント配線板を形成する場合に特に好適である。
【００２２】
ビルドアップの方法は問わないが、セミアディティブ法、フルアディティブ法等の公知のアディティブ法や、ガラスクロス、耐熱樹脂製不織布、金属箔等を芯材にして樹脂絶縁層をプリプレグでラミネートするラミネート法等を用いることができる。アディティブ法を用いる場合は、ビルドアップ層を形成する前には、露出した配線表面に化学エッチング、黒化処理、針状メッキ等の公知の粗化面形成処理を行う。それによって、続いて形成される樹脂絶縁層の密着力が向上できる。形成された樹脂絶縁層の表面を過マンガン酸カリウム処理、クロム酸処理等により粗化処理する。続いて形成される導体層の密着力を上げるためである導体層は無電解メッキ、電解メッキ、フォトリソグラフィを用いた公知の方法で形成できる。
【００２３】
本発明に用いるスルーホール用充填ペーストには、エポキシ樹脂と金属フィラーと硬化剤とを含む組成物をベースとする。所望の特性を得るために、この組成物に対して感光性樹脂、熱可塑性樹脂、無機フィラー、超微粒子無機フィラー、脱泡剤等を適宜選択して添加した組成物とすることができる。
【００２４】
エポキシ樹脂は、一般に硬化収縮が少ないため、硬化後の凹みを効果的に抑制するために用いられる。特には、耐熱性、耐湿性、耐薬品性に優れる芳香族エポキシ樹脂を用いることが望ましい。更には、ペーストを無溶剤化する為に、常温で液状のエポキシ樹脂（ＢＰＡ型、ＢＰＦ型等のＢＰ型やＰＮ型）を用いるのがよい。
【００２５】
金属フィラーは、硬化収縮をさらに低減したり、熱膨張率やペーストの粘度を制御したり、スルーホール充填材に導電性を付与するために用いられる。また、金属フィラーの添加によって、スルーホール充填材上に形成した銅メッキとの密着性を効果的に向上することができる。不純物酸素を若干含んだアトマイズ粉が安価でよい。金属フィラーとしては、銅、ニッケル等の卑金属単体、又はこれらから選ばれる卑金属と他の金属との合金のみならず、これらの金属や合金で被覆した無機フィラー（本発明では、この金属被覆した無機フィラーも金属フィラーの概念に含める。）等を用いることができる。尚、合金粉末の場合は、Ａｇ、Ａｕ、Ｐｔ等の貴金属との合金がよい。
【００２６】
この金属フィラーは、銅粉末、ニッケル粉末等の卑金属粉末がよい。不純物酸素を若干含んだアトマイズ粉が安価でよい。貴金属を用いるよりも安価だからである。特には銅粉末又は銅合金粉末がよい。銅粉末又は銅合金粉末であると、スルーホール充填材上に形成するメッキを、通常の銅メッキ工程と同様の工程で行えるからである。銅合金粉末の場合は、Ａｇ、Ａｕ、Ｐｔ等の貴金属との合金がよい。
【００２７】
金属フィラーの形状には、球状、鱗片状、樹枝状等があるが、特には球状が望ましい。ペーストの流動性が向上して、スルーホール内への充填性に優れるからである。金属フィラーの平均粒径としては、２０μｍ以下が好ましい。この平均粒径が２０μｍよりも大きいと、銅メッキとの密着性が低下するからである。更に好ましくは１〜１５μｍ、特には３〜１０μｍである。銅メッキのフクレの発生を効果的に防止できるからである。また、スルーホール充填性を良好にするために、見かけ密度が２ｇ／ｃｍ³以上の金属フィラーを用いるのがよい。
【００２８】
本発明のスルーホール充填用ペーストは、熱膨張係数等の物性値調整のために、超微粒子無機フィラーを含む。超微粒子無機フィラーを構成するセラミックとしては、酸化物（アルミナ、シリカ等）、窒化物（窒化珪素、チタンナイトライド等）、ケイ化物（タングステンシリサイド、チタンシリサイド等）、炭化物（タングステンカーバイド、チタンカーバイド等）等を用いることができる。特には、タイミクロン等の微細アルミナ粉末やアエロジル等の微細シリカ粉末がよい。また必要に応じて、消泡剤、レベリング剤等の微量添加剤を添加しても良い。
【００２９】
本発明のスルーホール充填用ペーストは、超微粒子無機フィラーを含むことにより、スルーホール充填用ペーストの粘度の経時変化を抑えることができる。さらに分散剤を添加しても良い。超微粒子無機フィラーとしては、超微粒子状のセラミック粉末がよい。ここにいう超微粒子とは、以下のような特性を有する粒子をいう。（１）粒子のＢＥＴ法による比表面積が４０〜４００ｍ^２／ｇである。ＢＥＴ法は周知の手法のため、その詳細はここでは割愛する。スルーホール充填用ペーストのスルーホールへの穴埋め時の流動性を良好に保ちながら経時変化による粘度変動を効果的に防止することができる。粒子のＢＥＴ法による比表面積の好ましい範囲は６０〜１００ｍ^２／ｇである。スルーホールへの穴埋め時に問題となる硬化後のスルーホール充填用ペーストにおける樹脂塊やボイドの発生を効果的に抑えることができる。（２）粒子の１次粒子径が５〜５０ｎｍである。１次粒子径がｎｍ（ナノメートル）オーダーの超微粒子を用いることで、スルーホール充填用ペーストにおける金属フィラーの沈降や粘度の経時変化を効果的に抑制することができる。粒子の１次粒子径の好ましい範囲は１０〜２０ｎｍである。スルーホール充填用ペーストの粘度を過度に増大させることがない。（３）粒子の見かけ比重が３０〜２００ｇ／ｌである。スルーホール充填用ペーストにおける金属フィラーの沈降防止や粘度の経時変化の抑制に効果的である。
【００３０】
超微粒子無機フィラーとしては、その４％分散液（水：メタノール＝１：１溶液中）のｐＨ値が３．５〜６．５であるとよい。ｐＨ値を中性から酸性よりに調整することで、塩基性のイミダゾール系化合物である硬化剤の室温下での活性を落として、使用中に発生するスルーホール充填用ペーストの硬化の進行による増粘を効果的に抑制できる。ただし、ｐＨ値が３．５よりも小さくなると、逆に塩基性のイミダゾール系化合物である硬化剤の室温下での活性が落ちすぎて、硬化不十分になるので好ましくない。また、ｐＨ値が６．５よりも大きくなると、粘度が逆に不安定になりため好ましくない。超微粒子無機フィラーの４％分散液（水：メタノール＝１：１溶液中）のｐＨ値は、例えば以下のように測定する。ここでは超微粒子無機フィラーとして微細シリカ（アエロジル）を用いて説明する。アエロジル４ｇを２００ｍｌビーカーにとる。そこにメタノール５０ｍｌを加えて攪拌する。さらに純粋５０ｍｌを加えて攪拌して、よく分散させる。この分散液にガラス電極を入れてｐＨ値を測定する。５分間のｐＨ値の変動が±０．０１以内となるまで放置して、そのときの値を超微粒子無機フィラーのｐＨ値として読み取る。
【００３１】
超微粒子無機フィラーに含まれるカーボン含有量は３〜５質量％であるとよい。超微粒子のエポキシ樹脂との濡れ性を良好にして、優れた分散性を得ることができる。その理由の詳細は不明であるが、超微粒子無機フィラーの表面に付着したカーボン成分が、超微粒子無機フィラー同士の分子間力等により凝集するのを防止するものと推察される。また、カーボンが静電気を逃がして、超微粒子が静電力で凝集するのを防止しているとも推察される。
【００３２】
超微粒子無機フィラーに含まれる水分含有量は０．０５〜０．１５質量％であるとよい。スルーホール充填ペーストの製造ロット毎の粘度を安定して保つことができる。水分含有量が０．０５質量％未満であると、雰囲気の湿度変化に対して粘度が不安定になるため好ましくない。また、水分含有量が０．１５質量％を超えると、製造初期の粘度変動が大きくなるため好ましくない。
【００３３】
超微粒子無機フィラーは、有機成分を除く無機成分全体における主たる無機成分の純度が９９．５％以上のものがよい。これは、カーボン等の有機成分を除いた純粋な無機成分全体を１００質量％ととしたときの純度をいう。超微粒子無機フィラーは、その微細さゆえに、スルーホール充填用ペースト全体に均一に分散している。そのため、不要な不純物を含むと、不純物に起因する悪影響（例えば、金属イオンのマイグレーション等）が、スルーホール全体に及ぶことになる。それを回避するためには、無機成分の純度を９９．５％以上にするのがよい。
【００３４】
本発明のスルーホール充填用ペーストを用いたプリント配線板は、スルーホールの信頼性に優れる利点を有するため、ビルドアップ工法を用いた多層プリント配線板用途に好適である。Ｖｉａ ｏｎ ＰＴＨ構造を採った場合においても、安定した接続信頼性が得られる利点がある。また、スルーホール開口端面近傍に凹凸や隙間を発生させないため、ＦＣ−ＰＧＡ型の多層プリント配線板のようにスルーホール直下のビルドアップ層にピン付けする場合においても、ピンの傾きや接合不良等の不具合の発生を防止できる利点がある。
【００３５】
上記スルーホール充填用ペーストは、プリント配線板を構成するコア基板の表面やスルーホールの壁面に形成される導体層が、水に対する接触角が９０度以上になるように疎水化処理されている場合に特に有用である。尚、未処理の場合の接触角は通常７０〜８０度、酸化処理（いわゆる黒化処理）の場合の接触角は通常７５〜９０度である。
【００３６】
本発明のプリント配線板に用いるコア基板の導体層（主に銅）は、絶縁層やソルダーレジストとの密着性の向上及び発錆防止のために、カップリング剤や防錆剤で導体層表面を疎水性つまり親油性に処理（特には、水に対する接触角（２θ）が９０度を超えるように疎水化処理）するとよい。これらの処理をすることで、導体層の表面と親油性の樹脂（例えばエポキシ樹脂）との濡れ性は良くなり（逆に水との濡れ性は悪くなる）、絶縁層やソルダーレジストの密着性は向上する。密着性の向上のためには、水に対する接触角が９５度以上がよく、特には１００度以上がよい。ここにいう接触角（２θ）は、水滴（８）の導体表面（７）との接触点（８０）から水滴の頂点（８１）へ引いた仮想線（８２）と、水平線（８３）とが形成する角度（θ）（９）の２倍角（２θ）をいう（図５を参照。）。
【００３７】
ところが一般には、導体層の表面が親油性となって樹脂との濡れ性が高くなるにつれてブリードアウト（熱硬化時の樹脂流れ）は増大し、スルーホール充填用ペーストの穴埋め性は低下してしまう。しかし、本発明のスルーホール充填用ペーストは、水に対する接触角が９０度以上になるように疎水化処理された（つまり、親油性の高い）導体層を有するコア基板へも良好な穴埋めをすることができる利点がある。
【００３８】
本発明のプリント配線板に用いる導体層の表面の少なくとも一部は、粗度Ｒzが０．３〜２０μｍ、好ましくは０．５〜１０μｍ、より好ましくは１〜５μｍとなっているのがよい。特には、少なくともスルーホール充填用ペーストが接触する部分が粗化されているのがよい。例えば、スルーホールの内壁面に形成されたスルーホール導体層の部分である。スルーホール充填用ペーストが導体層の表面の凹凸に食い込んで、密着性を向上させるアンカー効果を奏するからである。表面粗度Ｒzの制御については、特に制約は無く、マイクロエッチング法や黒化処理等の公知の方法で行えばよい。
【００３９】
【実施例】
以下に本発明の一態様を実施例を用いて説明する。尚、本発明は以下の実施例のみに限定されるものではない。
・実施例１；スルーホール充填用ペーストの評価
▲１▼スルーホール充填用ペーストの調製
表１の組成になるように、以下に示すエポキシ樹脂、イミダゾール系硬化剤、銅およびシリカフィラーを混合し、３本ロールミルで十分に分散させ、スルーホール充填用ペーストを調製する。
【００４０】
エポキシ樹脂は以下の３種類。
・商品名：ＹＬ９８０：ビスフェノールA型エポキシ樹脂（油化シェル製）
・商品名：Ｅ８２８：ビスフェノールA型エポキシ樹脂（油化シェル製）
・商品名：Ｅ１５２：フェノールノボラック型エポキシ樹脂（油化シェル製）
【００４１】
硬化剤は以下の４種類。
・商品名：２Ｐ４ＭＨＺ：２−フェニル−４−メチル−５−ヒドロキシメチル−イミダゾール（四国化成製）
・商品名：２ＰＨＺ：２−フェニル−４，５−ジヒドロキシメチル−イミダゾール（四国化成製）
・商品名：２Ｐ４ＭＺ：２−フェニル−４−メチル−イミダゾール（四国化成製）
・商品名：２ＰＺ−ＯＫ：２−フェニル−イミダゾール・イソシアヌル酸付加体（四国化成製）
【００４２】
無機フィラーは以下の４種類。ここで、「ＲＹ２００Ｓ：微細シリカ（日本アエロジル製）」は超微粒子無機フィラーである。
・商品名：ＳＦＲ−ＣＵ−５：球状銅フィラー，平均粒径５μｍ（日本アトマイズ加工製）
・商品名：ＳＦＲ−ＣＵ−１０：球状銅フィラー，平均粒径１０μｍ（日本アトマイズ加工製）
・商品名：ＳＯＣ２：球状シリカフィラー，平均粒径０．５μｍ（龍森製）
・商品名：ＲＹ２００Ｓ：微細シリカ（日本アエロジル製）、ＢＥＴ法による比表面積７６ｍ²／ｇ、１次粒子径（平均値）１６ｎｍ、シリカ純度９９．９％以上（無機成分全体における主たる無機成分であるシリカの純度を示す。）、見かけ比重５０ｇ／ｌ、水分含有量０．０８９質量％、カーボン含有量４．３質量％、４％分散液（水：メタノール＝１：１溶液中）のｐＨ値５．７２
【００４３】
▲２▼コア基板の疎水化処理
基板（２）は、厚み８００μｍのＢＴ（ビスマレイミド・トリアジン）樹脂製銅張基板を用いる。機械ドリル加工で孔開けした後、孔の内壁面（３）に厚み１５μｍの銅メッキにより導体層（４）を形成して、直径３００μｍのスルーホールを有するコア基板（１）を作製する（図２を参照。）。得られた基板の導体層の表面を、市販のマイクロエッチング液（商品名：MEC etch BOND、メック社製）を用いて粗面化処理して、粗化面（５）を形成する（図３を参照。）。その後、市販の防錆剤によって防錆処理（商標名：ＣＺ処理、メック社製）を施して疎水化面（６）を形成して、疎水化処理を完了する（図４を参照。）。
【００４４】
疎水化処理を施した導体層表面の水に対する接触角（２θ）を、接触角測定器（商品名：ＣＡ−Ａ、協和科学製）により液適法で測定したところ、接触角（２θ）は１０１度であった。ここにいう接触角（２θ）は、水滴（８）の導体表面（７）との接触点（８０）から水滴の頂点（８１）へ引いた仮想線（８２）と、水平線（８３）とが形成する角度（θ）（９）の２倍角（２θ）をいう（図５を参照。）。
【００４５】
▲３▼穴埋め試験
表１に示すそれぞれのスルーホール充填用ペーストをコア基板（１）のスルーホール内に充填する。真空吸引装置の付いた台座の上に不繊紙を設置し、上記基板を、台座の上に配置する。その上にスルーホールの位置に対応するように２００μｍ、３００μｍ、４００μｍの３種類の貫通孔を有するステンレス製の穴埋めマスクを設置する。次いで、表１の各穴埋め充填ペーストを載せ、ローラー式スキージを加圧しながら穴埋め充填を行う。その後、１２０℃×４０分の条件で仮キュアした後、１５０℃×５時間の条件でキュアして熱硬化する。ここで、硬化後のスルーホール充填材（１１）の表面に２０μｍを越える凹み（ｄ）（１０）が発生したものを不合格（図６を参照。）として、穴埋め性を評価する。スルーホール１００個中の合格率が８５％以上のものを合格と判断する。合否を表１に併記する。
【００４６】
▲４▼隙間観察
コア基板のスルーホール部を切断し、硬化後のスルーホール充填材の隙間の有無を倍率２００倍の拡大顕微鏡を用いて観察する。スルーホール１００個中の合格率が８５％以上のものを合格と判断する。合否を表１に併記する。
【００４７】
【表１】

【００４８】
一般式（１）の構造を持つイミダゾール系硬化剤を用いた実施例である試料番号１〜試料番号３は、隙間が発生せず、良好な結果が得られる。特に、Ｒ₁がメチル基である試料番号１および試料番号２では、凹みも殆ど発生せず（１０μｍ以下）、良好な穴埋めができた。
【００４９】
一方、分子内に水酸基を持たないイミダゾール系硬化剤を用いた比較例である試料番号４及び試料番号５では、内部に隙間が発生した。尚、参考例である試料番号６は、比較例である試料番号４の銅フィラーをシリカフィラーに変更したものであるが、これには隙間は発生しなかった。このことから、この隙間は、銅フィラーを用いた場合に固有の現象であることがわかる。
【００５０】
・実施例２；ＦＣ−ＰＧＡ型多層プリント配線板の例
▲１▼ＦＣ−ＰＧＡ型多層プリント配線板を用いた半導体装置の作製
本発明のプリント配線板の製造方法を用いた多層プリント配線板について、図７に示すいわゆる「ＦＣ−ＰＧＡ」構造の実施例を用いて以下に説明する。
図８に示すような、厚み０．４ｍｍの絶縁基板（１０１）に厚み１８μｍの銅箔（２００）を貼り付けたＦＲ−５製両面銅張りコア基板を用意する。ここで用いるコア基板の特性は、ＴＭＡによるＴｇ（ガラス転移点）が１７５℃、基板面方向のＣＴＥ（熱膨張係数）が１６ｐｐｍ／℃、基板面垂直方向のＣＴＥ（熱膨張係数）が５０ｐｐｍ／℃、１ＭＨｚにおける誘電率εが４．７、１ＭＨｚにおけるｔａｎδが０．０１８である。
【００５１】
コア基板上にフォトレジストフィルムを貼り付けて露光現像を行い、直径６００μｍの開口部及び所定の配線形状に対応する開口部（図示せず。）を設ける。フォトレジストフィルムの開口部に露出した銅箔を亜硫酸ナトリウムと硫酸を含むエッチング液を用いてエッチング除去する。フォトレジストフィルムを剥離除去して、図９に示すような露出部（３００）及び所定の配線形状に対応する露出部（図示せず。）が形成されたコア基板を得る。
【００５２】
市販のエッチング処理装置（メック社製 ＣＺ処理装置）によってエッチング処理を施して銅箔の表面粗化をした後、エポキシ樹脂を主体とする厚み３５μｍの絶縁フィルムをコア基板の両面に貼り付ける。そして、１７０℃×１．５時間の条件にてキュアして絶縁層を形成する。このキュア後の絶縁層の特性は、ＴＭＡによるＴｇ（ガラス転移点）が１５５℃、ＤＭＡによるＴｇ（ガラス転移点）が２０４℃、ＣＴＥ（熱膨張係数）が６６ｐｐｍ／℃、１ＭＨｚにおける誘電率εが３．７、１ＭＨｚにおけるｔａｎδが０．０３３、３００℃での重量減が−０．１％、吸水率が０．８％、吸湿率が１％、ヤング率が３ＧＨｚ、引っ張り強度が６３ＭＰａ、伸び率が４．６％である。
【００５３】
図１０に示すように、炭酸ガスレーザを用いて絶縁層（４００）に層間接続用のビアホール（５００）を形成する。ビアホールの形態は、表層部の直径は１２０μｍ、底部の直径は６０μｍのすりばち状である。更に炭酸ガスレーザの出力を上げて、絶縁層とコア基板を貫通するようにスルーホール（６００）を形成する。スルーホールの直径は、２００μｍ、３００μｍ、４００μｍの３種類（図示せず。）を形成する。これらのスルーホールの内壁面はレーザ加工に特有のうねり（図示せず。）を有する。そして、基板を塩化パラジウムを含む触媒活性化液に浸漬した後、全面に無電解銅メッキを施す（図示せず。）。
【００５４】
次いで、基板の全面に厚み１８μｍの銅パネルメッキ（７００）をかける。ここで、ビアホールには、層間を電気的に接続するビアホール導体（８００）が形成される。またスルーホールには、基板の表裏面を電気的に接続するスルーホール導体（９００）が形成される。パネルメッキ後のスルーホールの直径は、１６０μｍ、２６０μｍ、３６０μｍとなる。市販のエッチング処理装置（メック社製 ＣＺ処理装置）によってエッチング処理を施して銅メッキの表面粗化する。その後、同社の防錆剤によって防錆処理（商標名：ＣＺ処理）を施して疎水化面を形成して、疎水化処理を完了する。疎水化処理を施した導体層表面の水に対する接触角（２θ）を、接触角測定器（商品名：ＣＡ−Ａ、協和科学製）により液適法で測定したところ、接触角（２θ）は１０１度であった。
【００５５】
真空吸引装置の付いた台座の上に不繊紙を設置し、上記基板を、台座の上に配置する（図示せず。）。その上にスルーホールの位置に対応するように２００μｍ、３００μｍ、４００μｍの３種類の貫通孔を有するステンレス製の穴埋めマスクを設置する（図示せず。）。次いで、表１の各スルーホール充填用ペーストを載せ、ローラー式スキージを加圧しながら穴埋め充填を行う（図示せず。）。
【００５６】
図１１に示すように、スルーホール内に充填したスルーホール充填用ペースト（１０２）を、１２０℃×２０分の条件下で仮キュアさせる。次いで、図１２に示すように、ベルトサンダー（粗研磨）を用いてコア基板表面を研磨した後、バフ研磨（仕上げ研磨）して平坦化して、１５０℃×５時間の条件下でキュアさせて、穴埋め工程を完了する。尚、この穴埋め工程を完了した基板の一部は、穴埋め性の評価試験に用いる。
【００５７】
図１３に示すように、金型（図示せず。）を用いて□８ｍｍの貫通孔（１１０）を形成する。図１４に示すように、基板の一面にマスキングテープ（１２０）を貼り付ける。そして、図１５に示すように、貫通孔（１１０）に露出したマスキングテープ（１２０）の粘着面上に、積層チップコンデンサ（１３０）をチップマウンタを用いて８個配置する。この積層チップコンデンサは、絶縁体と導体とを交互に積層した積層体（１５０）からなり、電極（１４０）が積層体から７０μｍ突き出している。
【００５８】
図１６に示すように、積層チップコンデンサを配置した貫通孔の中に、エポキシ樹脂とシリカフィラーを主体とする埋め込み樹脂（１６０）をディスペンサ（図示せず。）を用いて充填する。埋め込み樹脂を、１次加熱工程を８０℃×３時間、２次加熱工程を１７０℃×６時間の条件により脱泡および熱硬化する。
【００５９】
図１７に示すように、硬化した埋め込み樹脂の表面を、ベルトサンダーを用いて粗研磨した後、ラップ研磨にて仕上げ研磨する。研磨面には、チップコンデンサーの電極の端面が露出している。次いで、仮キュアした穴埋め樹脂を１５０℃×５時間の条件下で硬化させる。
【００６０】
その後、膨潤液とＫＭｎＯ₄溶液を用いて、埋め込み樹脂の研磨面を粗化する。粗化面をＰｄ触媒活性化した後、無電解メッキ、電解メッキの順番で銅メッキを施す。図１８に示すように、埋め込み樹脂の上に形成されたメッキ層は、チップコンデンサーの電極の端面と電気的に接続されている。メッキ面にレジストを形成し、所定の配線パターンをパターニングする。不要な銅をＮａ₂Ｓ₂Ｏ₈／濃硫酸を用いてエッチング除去する。レジストを剥離して、図１９に示すように、配線の形成を完了する。市販のエッチング処理装置（メック社製 ＣＺ処理装置）によってエッチング処理を施して配線の銅メッキの表面粗化する。
【００６１】
その上に絶縁層となるフィルム（１９０）をラミネートして熱硬化した後、炭酸ガスレーザーを照射して層間接続用のビアホールを形成する。絶縁層の表面を上記と同じ酸化剤を用いて粗化し、同様の手法で所定の配線（２００）を形成する。配線基板の最表面にソルダーレジスト層となるドライフィルムをラミネートして、半導体素子の実装パターンを露光、現像して形成して、ソルダーレジスト層（２１０）の形成を完了する。実装用のピン付けを行う面についても同様の方法により、所定の配線（２３０）とソルダーレジスト層（２４０）を形成して、図２０に示すように、ピン付け前の多層プリント配線板を得る。
【００６２】
半導体素子を実装する端子電極（２００）には、Ｎｉメッキ、Ａｕメッキの順番でメッキを施す（図示せず。）。その上に低融点ハンダからなるハンダペーストを印刷した後、ハンダリフロー炉を通して半導体素子を実装するためのハンダバンプ（２２０）を形成する。
【００６３】
一方、半導体素子実装面の反対側には、高融点ハンダからなるハンダペーストを印刷した後、ハンダリフロー炉を通してピン付けするためのハンダバンプ（２６０）を形成する。治具（図示せず。）にピン（２５０）をセットした上に基板を配置した状態で、ハンダリフロー炉を通してピン付けを行い（図示せず。）、図２１に示すように、半導体素子を実装する前のＦＣ−ＰＧＡ型の多層プリント配線板を得る。投影機を用いてピンの先端の所定位置からの位置ずれ量を測定したところ、０．１ｍｍ以下と良好な結果が得られた。
【００６４】
半導体素子実装面上に半導体素子（２７０）を実装可能な位置に配置して、低融点ハンダのみが溶解する温度条件にてハンダリフロー炉を通して、半導体素子を実装する。実装部にアンダーフィル材をディスペンサーで充填した後、熱硬化して、図７示すような半導体素子を実装したＦＣ−ＰＧＡ型の多層プリント配線板を用いた半導体装置を得る。
【００６５】
▲２▼スルーホール充填性の評価
上記の穴埋め工程を完了した基板を用いて、穴埋め性の評価試験を行う。スルーホール充填材のスルーホール端面から倍率２００倍の拡大鏡を用いて、クラック等の不具合の有無を目視検査する。
【００６６】
一方、比較例は、穴埋め工程がメタルマスクを用いた公知のスクリーン印刷法であり、表２に示す条件で行う。メタルマスクの貫通孔の直径は実施例の支持板と同様にする。充填終了後、スルーホール端面から倍率２００倍の拡大鏡を用いて、スルーホール充填材の不具合の有無を目視検査する。結果を表２に示す。また、硬化後のスルーホール充填材のスルーホール端面から倍率２００倍の拡大鏡を用いて、クラック等の不具合の有無を目視検査する。
【００６７】
【表２】

【００６８】
表２の結果より、本発明によれば、コア基板の両面にビルドアップ層を形成した基板を貫通するように形成され、かつ、その内壁面にうねりを有するスルーホールであっても、充填不良や硬化時の隙間等の不具合を生じさせることなく充填、硬化が可能であることがわかる。１枚の基板中に異なる直径のスルーホールを有する場合であっても、充填性を均一にできる。また、スルーホール開口端面近傍に凹凸や隙間を発生させないため、続いて形成されるビルドアップ層に凹凸を生じさせることもない。そのため、ＦＣ−ＰＧＡ型の配線基板のようにスルーホール直下のビルドアップ層にピン付けした多層プリント配線板に好適である。
【００６９】
・実施例３；超微粒子無機粒子の物性によるスルーホール充填ペーストへの影響の評価
▲１▼スルーホール充填用ペーストの調整
諸特性値が表３に記載するように異なる以外は、組成は実施例１の試料番号１の組成を用いてスルーホール充填用ペーストを調整する。調整方法も実施例１に準じて行う。一部の試料は、アエロジルではなく、アルミナ、チタニアを用いる。
▲２▼評価
実施例１と同様の方法を用いて穴埋め試験を行う。結果は、穴埋め性の極めて良好なものを◎（ほぼ１００％合格）、穴埋め性のかなり良好なものを○（９５％以上１００％未満合格）、穴埋め性の良好なものを△（８５％以上９５％未満合格）として表３に示す。また、粘度の経時変化は、スルーホール充填用ペーストを作製してから２ヶ月放置したものを用いる。放置後の状態を確認して、増粘の殆ど無いものを◎、若干の増粘に留まるものを○、増粘しているものを△として表３に併せて示す。
【００７０】
【表３】

【００７１】
結果より、超微粒子無機フィラーの特性を特定の範囲に調整することで、穴埋め性のみならず、粘度の経時変化を効果的に抑制できることがわかる。超微粒子無機フィラーの特性値の臨界的意義について、以下に補足説明をする。
【００７２】
試料番号１３を見ると、比表面積がこれ以上大きくなると高粘度になりすぎて、穴埋め性によくないことがわかる。また、１次粒子径の下限は、せいぜい５ｎｍが許容範囲であることがわかる。試料番号２８を見ると、比表面積がこれ以上小さくなると粘度の経時変化が大きくなってよくないことがわかる。試料番号２６を見ると、１次粒子径がこれ以上になると、粘度の経時変化が大きくなってよくないことがわかる。
【００７３】
試料番号１４を見ると、ｐＨ値がこれ以上小さくなると粘度の経時変化が大きくなってよくないことがわかる。この理由の詳細は不明であるが、酸性よりになると、塩基性の硬化剤と好ましくない副反応が起きているものと推察される。また、試料番号１９を見ると、ｐＨ値がこれ以上大きくなると、若干粘度の経時変化が大きくなってよくないことがわかる。この理由の詳細は不明であるが、必要以上に塩基性に偏ると、何らかの好ましくない副反応が起きるものと推察される。
【００７４】
試料番号２１を見ると、これ以上カーボン含有量が少なくなると、穴埋め性が低下することがわかる。カーボン成分が分子間力等による凝集を抑えて、粘度変動を抑えるものと推察される。カーボン含有量が５質量％の試料番号２０では、試料番号２１のような特性低下は目立って無いものの、これ以上含めるとスルーホール上へのメッキに支障がでるため好ましくない。
【００７５】
試料番号２２を見ると、これ以上水分含有量が多くなると、粘度の経時変化が大きくなってよくないことがわかる。水分が樹脂の硬化性に影響を及ぼしているものと推察される。また、試料番号２３を見ると、これ以上水分含有量が少なくなっても、やはり若干粘度の経時変化が大きくなってよくないことがわかる。おそらく、作業環境の湿度を吸湿しやすくなっていることが影響しているものと思われる。
【００７６】
【発明の効果】
本発明によれば、銅等の卑金属フィラーを含有するスルーホール充填材において、耐熱性の低下や表裏面の凹みの発生を伴うことなく、硬化後のスルーホール充填材の内部に隙間が発生するのを防止することができる。また、スルーホール開口端面近傍に凹凸や隙間を発生させないため、続いて形成されるビルドアップ層に凹凸を生じさせることもない。スルーホール近傍のビルドアップ層の平坦性に優れた基板を得ることができるため、ＦＣ−ＰＧＡ型の配線基板のようにスルーホール直下のビルドアップ層にピン付けしたＰＧＡ（ピングリッドアレイ）型の多層プリント配線板を形成する場合に特に好適である。
【図面の簡単な説明】
【図１】硬化後のスルーホール充填材の内部に発生した隙間を示す説明図。
【図２】コア基板の疎水化処理の工程を示す説明図。
【図３】コア基板の疎水化処理の工程を示す説明図。
【図４】コア基板の疎水化処理の工程を示す説明図。
【図５】導体表面の水に対する接触角θの測定方法を示す説明図。
【図６】硬化後のスルーホール充填材の表面に発生する凹みｄを示す説明図。
【図７】本発明の実施例で用いたＦＣ−ＰＧＡ型の多層プリント配線板を用いた半導体装置の説明図。
【図８】本発明の実施例で用いた厚み４００μｍの銅張りコア基板の概略図。
【図９】厚み４００μｍの銅張りコア基板のパターニング後の状態を示す説明図。
【図１０】コア基板の両面に絶縁層を形成した基板にビアホールとスルーホールを形成した状態を示す説明図。
【図１１】コア基板の両面に絶縁層を形成した基板にパネルメッキをかけた後の状態を示す説明図。
【図１２】スルーホールを穴埋め充填した基板の説明図。
【図１３】貫通孔を打ち抜き形成した基板を示す説明図。
【図１４】貫通孔を打ち抜き形成した基板の一面にマスキングテープを貼り付けた状態を示す説明図。
【図１５】貫通孔内に露出したマスキングテープ上に積層チップコンデンサを配置した状態を示す説明図。
【図１６】貫通孔内に埋め込み樹脂を充填した状態を示す説明図。
【図１７】基板面を研磨して平坦化した状態を示す説明図。
【図１８】基板の研磨面にパネルメッキをかけた状態を示す説明図。
【図１９】配線をハターニングした状態を示す説明図。
【図２０】基板上にビルドアップ層及びソルダーレジスト層を形成した状態を示す説明図。
【図２１】本発明の実施例で用いたＦＣ−ＰＧＡ型の多層プリント配線板の説明図。
【符号の説明】
１ コア基板
２ 基板
３ 孔の内壁面
４ 導体層
５ 粗化面
６ 疎水化面
７ 導体層表面
８ 水滴
８０ 水滴の導体層表面との接触点
８１ 水滴の頂点
８２ 仮想線
８３ 水平線
９ 角度（θ）
１０ 硬化後のスルーホール充填材
１１ 凹み（ｄ）
１２ 隙間[0001]
[Industrial application fields]
The present invention relates to a through-hole filling paste used by filling a through-hole formed in a printed wiring board or a laminated board.
[0002]
[Prior art]
In recent years, in order to further increase the density of a printed circuit board, a structure (Via on PTH) in which a via hole for electrically connecting an insulating layer is formed immediately above a through hole has been studied (Japanese Patent Laid-Open No. Hei 6). (See -275959). In this structure, it is necessary to perform copper plating on the through hole filler filled in the through hole. In order to obtain high connection reliability, it is important to ensure the adhesion of copper plating to the through-hole filler.
[0003]
When using a paste for filling through-holes containing only inorganic fillers such as silica, a special chemical treatment using chromic acid or potassium permanganate is applied to improve the adhesion of copper plating. It is necessary to roughen the surface of the through-hole filler. For this reason, it is difficult to reduce the manufacturing cost.
[0004]
Then, in order to raise adhesiveness with copper plating, use of the paste for through-hole filling containing a metal filler like copper is examined (refer Unexamined-Japanese-Patent No. 11-140280).
[0005]
[Problems to be solved by the invention]
However, when a through-hole filling paste containing a copper filler is used, a crack-like gap (12) may be generated inside the cured through-hole filler (100) (FIG. 1). See). Such a crack-like gap does not occur inside a through-hole filler containing only an inorganic filler, but tends to occur inside a through-hole filler containing a metal filler such as copper.
[0006]
This gap was considered to be a crack caused by internal stress generated by curing shrinkage of the resin component of the through-hole filler. Therefore, in order to prevent the occurrence of crack-like gaps, it has been studied to add a component that imparts flexibility to the through-hole filler after curing to reduce internal stress (Japanese Patent Application Laid-Open No. Hei 8). -311157.)
[0007]
However, the addition of a component that imparts flexibility causes a decrease in the glass transition point (Tg) of the through-hole filler after curing, thereby reducing heat resistance, or volume shrinkage (re-generation) that occurs in the solder reflow process. It is known to increase (curing shrinkage) (see JP-A-11-199759). Phenomena such as a decrease in heat resistance and re-curing shrinkage are undesirable factors that reduce the reliability of through-holes in printed wiring boards and multilayer printed wiring boards.
[0008]
Thus, since there is a problem such as a decrease in the heat resistance of the through-hole filler and the occurrence of dents on the front and back surfaces, flexibility is given to those that require high reliability such as MPU packages. It was not possible to use a paste for filling through-holes to which various components were added. Therefore, it is not easy to obtain a printed wiring board or a multilayer printed wiring board with high reliability related to the through hole. Further, since the through hole filling paste has a large viscosity variation due to the change with time, it is necessary to suppress this variation in order to obtain a stable product. Furthermore, it is necessary to ensure a long shell life so that stable characteristics can be maintained even after long-term storage.
[0009]
[Means for Solving the Problems]
The curing shrinkage behavior of the through-hole filling paste to which a metal filler containing a base metal such as copper is added is greatly different in the curing rate between the central portion and the surface portion (portion exposed to the outside air) of the through-hole. More specifically, the curing rate at the center of the through hole is extremely slow compared to the curing rate at the surface of the through hole.
[0010]
In view of these phenomena, the present inventors have found that the above-mentioned gap is not a crack caused by internal stress as conventionally described, but the resin component in the central portion having a low curing rate is formed on the surface portion having a high curing rate. It was inferred that it was a nest generated by moving (so-called blow hole).
[0011]
Therefore, the present inventors examined changes in curing behavior using various curing agents. As a result, the inventors have found that by selecting a specific curing agent, the curing shrinkage behavior of the paste for filling through-holes can be controlled and the occurrence of the gaps can be suppressed, and the present invention has been completed. It is.
[0012]
It is known that base metals such as copper used for the metal filler have a certain catalytic action and have some influence on the curing reaction of the epoxy resin (see Japanese Patent Laid-Open No. 9-31307). The problem of the curing shrinkage behavior seems to be due to the fact that the catalytic effect varies depending on the oxidation state of the base metal surface. Specifically, the anionic polymerization is inhibited on the copper surface, while the anionic polymerization is promoted on the oxidized surface.
[0013]
In the present invention, an imidazole-based curing agent having the structure of the general formula (1) is used. By using this hardening | curing agent, the problem of the said clearance gap which generate | occur | produces in the through-hole filler which added the metal filler containing base metals, such as copper powder, can be avoided. This imidazole-based curing agent is presumed to have an effect of adjusting the variation in the catalytic action due to the difference in the surface state of the base metal and promoting the homogenization of the curing shrinkage behavior. The reason for this will be described below.
[0014]
In general, it is said that an epoxy resin to which an imidazole-based curing agent is added progresses in polymerization while forming an intramolecular complex (Polymer Publications, "Introduction Epoxy Resin" p.98-p.99, reference.). It is said that when an intramolecular complex is formed, the reaction rate decreases, but instead, a polymerization termination reaction hardly occurs. That is, it becomes difficult to be influenced from the outside which inhibits the polymerization reaction of the epoxy resin. The present inventors have inferred that in order to promote the homogenization of the curing shrinkage behavior, the point is how to effectively form this intramolecular complex. Therefore, the inventors have investigated using various imidazole-based curing agents, and found that by selecting an imidazole-based curing agent having a specific structure, it is possible to promote the homogenization of the curing shrinkage behavior of the epoxy resin. The invention has been completed.
[0015]
The imidazole curing agent of the general formula (1) used in the present invention has a hydroxyl group (—OH) in the molecule. An intramolecular complex is more effectively formed by hydrogen bonding between the hydroxyl group and the anionic polymerization terminal group. As a result, it becomes difficult to be affected by variations in the catalytic action of the base metal, it is possible to promote uniformization of the curing shrinkage behavior, and it is presumed that the generation of the gap can be effectively suppressed.
[0016]
General formula (1)
[Chemical 1]

[0017]
R in the general formula (1) ₁ Examples of preferable substituents include a hydrogen atom, an alkyl group having 1 to 10 carbon atoms, a hydroxyalkyl group having 1 to 10 carbon atoms, and an alkyloxy group having 1 to 10 carbon atoms. Furthermore, a hydrogen atom, an alkyl group having 1 to 4 carbon atoms (methyl group, ethyl group, propyl group, isopropyl group, butyl group), a hydroxyalkyl group having 1 to 4 carbon atoms (hydroxymethyl group, hydroxyethyl group, hydroxy group) Propyl group, hydroxybutyl group) or an alkyloxy group having 1 to 4 carbon atoms (methoxy group, ethoxy group, propoxy group, isopropoxy group, butoxy group) is preferable.
[0018]
Particularly preferred imidazole curing agents are R ₁ Is a methyl group, “2-phenyl-4-methyl-5-hydroxymethyl-imidazole”. This curing agent can effectively suppress the above-mentioned gap problem that is likely to occur in the through-hole filler to which copper powder is added. In addition, it is possible to effectively suppress the problem of dents on the front and back surfaces of the through-hole filler, which occurs when the through-hole filling paste filled in the through holes is cured. If dents occur on the front and back surfaces of the through-hole filler, it is not preferable because it causes unevenness in the build-up layer to be formed next. In particular, there is a problem even when pinning to a build-up layer immediately below a through-hole like an FC-PGA type wiring board as shown in FIG. The “FC-PGA type wiring board” here refers to a PGA (pin grid array) type wiring board of a type in which a semiconductor element is mounted by FC (flip chip) and mounted on a mother board by pins.
[0019]
In particular, a through-hole having a large aspect ratio (1.6 to 32, preferably 2.5 to 32, more preferably 5 to 32, particularly 8 to 32), or etching or black on the inner wall surface of the through-hole conductor. The roughened surface having irregularities with a roughness Rz of 0.3 to 3 μm (preferably 2 to 3 μm) due to ten-point average roughness (according to 3.5.1 of JIS B 0601) is formed by the treatment. Or when there are undulations or steps on the inner wall surface and through holes with poor linearity (such as those drilled by a laser) or through holes with different diameters (for example, 100 μm and 300 μm are mixed) This is useful when filling each through hole. A through-hole conductor for conduction is preferably formed on the inner wall surface of the through-hole. However, when a conductive through-hole filling paste is used, the formation of the through-hole conductor can be omitted. .
[0020]
As a substrate having a through hole used in the present invention, it is preferable to use a so-called core substrate such as FR-4, FR-5, or BT. However, a thick copper foil having a thickness of about 35 μm is formed on a thermoplastic resin sheet such as PTFE. You may use what formed the through hole in what was pinched | interposed and used as the core board | substrate. In addition, a substrate having a through hole formed so as to penetrate a substrate having an insulating layer formed on at least one surface of the core substrate can be used. FIG. 11 is an example of a substrate having through holes formed so as to penetrate through a substrate having an insulating layer formed on both surfaces of the core substrate. In this case, the through hole is preferably formed using a laser processing machine using a carbon dioxide gas laser or the like after the insulating layer is formed. However, through-holes drilled by a laser are often difficult to fill with through-holes due to undulations and steps on the inner wall surface and poor linearity. The through-hole filling paste used in the present invention has an advantage that even if it is a through-hole drilled by a laser so as to penetrate a substrate having an insulating layer formed on both sides of the core substrate, it can be satisfactorily filled. is there.
[0021]
A multilayer printed wiring board can be obtained by forming a build-up layer in which insulating layers and conductor layers are alternately laminated on the printed wiring board obtained by the method for producing a printed wiring board of the present invention. In this case, even with a multilayer printed wiring board with a built-in capacitor as shown in FIG. 7, the thickness of the so-called glass-epoxy composite material (insulating substrate) is reduced to about 400 μm, half of the usual 800 μm. There is an advantage that the height can be reduced. Further, since a substrate having no gap inside the through-hole filling material and excellent in the flatness of the build-up layer in the vicinity of the through-hole is used, the through-hole is formed like an FC-PGA type wiring substrate as shown in FIG. This is particularly suitable when forming a PGA (pin grid array) type multilayer printed wiring board that is pinned to a buildup layer immediately below the hole.
[0022]
The build-up method is not limited, but a known additive method such as a semi-additive method or a full additive method, or a laminating method in which a resin insulation layer is laminated with a prepreg using a glass cloth, a heat-resistant resin nonwoven fabric, a metal foil or the like as a core material. Etc. can be used. When the additive method is used, a known roughened surface forming process such as chemical etching, blackening process, or acicular plating is performed on the exposed wiring surface before forming the buildup layer. Thereby, the adhesive force of the resin insulating layer formed subsequently can be improved. The surface of the formed resin insulation layer is roughened by potassium permanganate treatment, chromic acid treatment or the like. The conductor layer for increasing the adhesion of the subsequently formed conductor layer can be formed by a known method using electroless plating, electrolytic plating, or photolithography.
[0023]
The filling paste for through holes used in the present invention is based on a composition containing an epoxy resin, a metal filler, and a curing agent. In order to obtain desired characteristics, a composition in which a photosensitive resin, a thermoplastic resin, an inorganic filler, an ultrafine inorganic filler, a defoaming agent, and the like are appropriately selected and added to the composition can be obtained.
[0024]
Epoxy resins generally have little cure shrinkage and are used to effectively suppress dents after cure. In particular, it is desirable to use an aromatic epoxy resin having excellent heat resistance, moisture resistance, and chemical resistance. Furthermore, in order to make the paste solvent-free, it is preferable to use an epoxy resin (BP type such as BPA type or BPF type or PN type) that is liquid at room temperature.
[0025]
The metal filler is used for further reducing curing shrinkage, controlling the coefficient of thermal expansion and the viscosity of the paste, and imparting conductivity to the through-hole filler. In addition, the addition of the metal filler can effectively improve the adhesion with the copper plating formed on the through-hole filler. Atomized powder containing some impurity oxygen may be inexpensive. Examples of the metal filler include not only a base metal such as copper and nickel, or an alloy of a base metal selected from these and other metals, but also an inorganic filler coated with these metals or alloys (in the present invention, this metal-coated inorganic filler). A filler is also included in the concept of a metal filler.) Etc. can be used. In the case of an alloy powder, an alloy with a noble metal such as Ag, Au, or Pt is preferable.
[0026]
The metal filler is preferably a base metal powder such as copper powder or nickel powder. Atomized powder containing some impurity oxygen may be inexpensive. This is because it is cheaper than using precious metals. In particular, copper powder or copper alloy powder is preferable. This is because when the copper powder or the copper alloy powder is used, the plating formed on the through-hole filler can be performed in the same process as the normal copper plating process. In the case of copper alloy powder, an alloy with a noble metal such as Ag, Au, or Pt is preferable.
[0027]
The metal filler has a spherical shape, a scale shape, a dendritic shape, etc., and a spherical shape is particularly desirable. This is because the fluidity of the paste is improved and the filling property into the through hole is excellent. As an average particle diameter of a metal filler, 20 micrometers or less are preferable. This is because when the average particle size is larger than 20 μm, the adhesion with the copper plating is lowered. More preferably, it is 1-15 micrometers, especially 3-10 micrometers. It is because generation | occurrence | production of the swelling of copper plating can be prevented effectively. Moreover, in order to improve the through hole filling property, the apparent density is 2 g / cm. ^Three The above metal filler is preferably used.
[0028]
The through-hole filling paste of the present invention is used for adjusting physical properties such as a coefficient of thermal expansion. Ultrafine particles Contains inorganic filler. Ultrafine particles Ceramics constituting the inorganic filler include oxides (alumina, silica, etc.), nitrides (silicon nitride, titanium nitride, etc.), silicides (tungsten silicide, titanium silicide, etc.), carbides (tungsten carbide, titanium carbide, etc.) Etc. can be used. In particular, fine alumina powder such as Tymicron and fine silica powder such as Aerosil are preferable. Moreover, you may add trace additives, such as an antifoamer and a leveling agent, as needed.
[0029]
The through-hole filling paste of the present invention includes an ultrafine inorganic filler. By Changes in the viscosity of the through-hole filling paste over time can be suppressed. Further, a dispersant may be added. As the ultrafine inorganic filler, ultrafine ceramic powder is preferable. The ultrafine particles referred to here are particles having the following characteristics. (1) Specific surface area of particles by BET method is 40-400m ² / G. Since the BET method is a well-known method, its details are omitted here. Viscosity fluctuations due to changes over time can be effectively prevented while maintaining good fluidity when filling the through-holes with the paste for filling through-holes. The preferable range of the specific surface area of the particles by the BET method is 60 to 100 m. ² / G. Generation | occurrence | production of the resin lump and void in the paste for through-hole filling after hardening which becomes a problem at the time of filling in a through-hole can be suppressed effectively. (2) The primary particle diameter of the particles is 5 to 50 nm. By using ultrafine particles having a primary particle size of the order of nm (nanometers), it is possible to effectively suppress sedimentation of metal fillers and changes in viscosity over time in the paste for filling through-holes. A preferable range of the primary particle diameter of the particles is 10 to 20 nm. The viscosity of the paste for filling through holes is not excessively increased. (3) The apparent specific gravity of the particles is 30 to 200 g / l. It is effective for preventing sedimentation of the metal filler in the through-hole filling paste and suppressing the change in viscosity over time.
[0030]
As the ultrafine inorganic filler, the pH value of the 4% dispersion (water: methanol = 1: 1 solution) is preferably 3.5 to 6.5. By adjusting the pH value from neutral to acidic, the activity of the curing agent, which is a basic imidazole compound, is reduced at room temperature, and the increase in the progress of curing of the through-hole filling paste that occurs during use is increased. Viscosity can be effectively suppressed. However, if the pH value is smaller than 3.5, the curing agent, which is a basic imidazole compound, is not preferable because the activity at room temperature is too low and the curing becomes insufficient. On the other hand, a pH value higher than 6.5 is not preferable because the viscosity becomes unstable. The pH value of a 4% dispersion of ultrafine inorganic filler (water: methanol = 1: 1 solution) is measured, for example, as follows. Here, description will be made using fine silica (Aerosil) as the ultrafine inorganic filler. Take 4 g of Aerosil in a 200 ml beaker. 50 ml of methanol is added and stirred. Add 50 ml of pure water and stir to disperse well. A glass electrode is put into this dispersion and the pH value is measured. The sample is left until the fluctuation of the pH value for 5 minutes is within ± 0.01, and the value at that time is read as the pH value of the ultrafine inorganic filler.
[0031]
The carbon content contained in the ultrafine inorganic filler is preferably 3 to 5% by mass. It is possible to improve the wettability of the ultrafine epoxy resin and to obtain excellent dispersibility. Although the details of the reason are unknown, it is presumed that the carbon component adhering to the surface of the ultrafine inorganic filler is prevented from aggregating due to the intermolecular force between the ultrafine inorganic fillers. It is also speculated that carbon releases static electricity and prevents the ultrafine particles from aggregating with electrostatic force.
[0032]
The water content contained in the ultrafine inorganic filler is preferably 0.05 to 0.15% by mass. The viscosity of each production lot of through-hole filling paste can be stably maintained. If the water content is less than 0.05% by mass, the viscosity becomes unstable with respect to the humidity change of the atmosphere, which is not preferable. On the other hand, when the water content exceeds 0.15% by mass, the viscosity fluctuation at the initial stage of production becomes large, which is not preferable.
[0033]
The ultrafine inorganic filler preferably has a purity of 99.5% or more of the main inorganic component in the whole inorganic component excluding the organic component. This means the purity when the total amount of pure inorganic components excluding organic components such as carbon is 100% by mass. Due to its fineness, the ultrafine inorganic filler is uniformly dispersed throughout the through-hole filling paste. For this reason, when unnecessary impurities are included, adverse effects caused by the impurities (for example, migration of metal ions, etc.) reach the entire through hole. In order to avoid this, the purity of the inorganic component is preferably 99.5% or more.
[0034]
Since the printed wiring board using the through-hole filling paste of the present invention has an advantage of excellent through-hole reliability, it is suitable for multilayer printed wiring board applications using a build-up method. Even when the Via on PTH structure is adopted, there is an advantage that stable connection reliability can be obtained. In addition, since unevenness and gaps are not generated near the end face of the through-hole opening, even when pinned to a build-up layer directly below the through-hole, such as an FC-PGA type multilayer printed wiring board, pin inclination, poor bonding, etc. This has the advantage of preventing the occurrence of defects.
[0035]
When the through-hole filling paste is hydrophobized so that the conductor layer formed on the surface of the core substrate constituting the printed wiring board and the wall surface of the through-hole has a contact angle with water of 90 degrees or more Is particularly useful. The contact angle in the case of untreated is usually 70 to 80 degrees, and the contact angle in the case of oxidation treatment (so-called blackening treatment) is usually 75 to 90 degrees.
[0036]
The conductor layer (mainly copper) of the core substrate used for the printed wiring board of the present invention is a surface of the conductor layer with a coupling agent or a rust preventive agent in order to improve adhesion to an insulating layer or a solder resist and to prevent rusting. May be treated with hydrophobicity, that is, lipophilicity (in particular, hydrophobization treatment so that the contact angle (2θ) with respect to water exceeds 90 degrees). By performing these treatments, the wettability between the surface of the conductor layer and the lipophilic resin (for example, epoxy resin) is improved (in contrast, the wettability with water is deteriorated), and the adhesion between the insulating layer and the solder resist is improved. Will improve. In order to improve the adhesion, the contact angle with water is preferably 95 degrees or more, and particularly preferably 100 degrees or more. The contact angle (2θ) here refers to the imaginary line (82) drawn from the contact point (80) with the conductor surface (7) of the water drop (8) to the apex (81) of the water drop and the horizontal line (83). This is the double angle (2θ) of the angle (θ) (9) to be formed (see FIG. 5).
[0037]
However, in general, as the surface of the conductor layer becomes oleophilic and the wettability with the resin increases, the bleed out (the resin flow at the time of thermosetting) increases and the filling ability of the paste for filling through holes decreases. . However, the through-hole filling paste of the present invention also fills a core substrate having a conductive layer hydrophobized (that is, highly lipophilic) so as to have a contact angle with water of 90 ° or more. There are advantages that can be made.
[0038]
At least a part of the surface of the conductor layer used in the printed wiring board of the present invention has a roughness Rz of 0.3 to 20 μm, preferably 0.5 to 10 μm, more preferably 1 to 5 μm. In particular, it is preferable that at least a portion in contact with the through-hole filling paste is roughened. For example, it is a portion of the through-hole conductor layer formed on the inner wall surface of the through-hole. This is because the through-hole filling paste has an anchor effect that bites into the irregularities on the surface of the conductor layer and improves adhesion. Control of the surface roughness Rz is not particularly limited, and may be performed by a known method such as a microetching method or a blackening treatment.
[0039]
【Example】
One embodiment of the present invention will be described below with reference to examples. In addition, this invention is not limited only to a following example.
Example 1: Evaluation of paste for filling through-holes
(1) Preparation of through-hole filling paste
The following epoxy resin, imidazole curing agent, copper and silica filler are mixed so as to have the composition shown in Table 1, and sufficiently dispersed by a three-roll mill to prepare a through-hole filling paste.
[0040]
There are three types of epoxy resin:
・ Product name: YL980: Bisphenol A type epoxy resin (manufactured by oil-based shell)
・ Product name: E828: Bisphenol A type epoxy resin (Oilized shell)
-Product name: E152: Phenol novolac epoxy resin (Oilized shell)
[0041]
The following four types of curing agents.
-Trade name: 2P4MHZ: 2-phenyl-4-methyl-5-hydroxymethyl-imidazole (manufactured by Shikoku Chemicals)
-Product name: 2PHZ: 2-phenyl-4,5-dihydroxymethyl-imidazole (manufactured by Shikoku Chemicals)
-Product name: 2P4MZ: 2-phenyl-4-methyl-imidazole (manufactured by Shikoku Chemicals)
-Product name: 2PZ-OK: 2-phenyl-imidazole-isocyanuric acid adduct (manufactured by Shikoku Chemicals)
[0042]
The following four types of inorganic fillers. Here, “RY200S: fine silica (manufactured by Nippon Aerosil Co., Ltd.)” is an ultrafine inorganic filler.
-Product name: SFR-CU-5: Spherical copper filler, average particle size 5 μm (manufactured by Nippon Atomizing)
-Product name: SFR-CU-10: Spherical copper filler, average particle size 10 μm (manufactured by Nippon Atomizing)
-Product name: SOC2: Spherical silica filler, average particle size 0.5 μm (manufactured by Tatsumori)
-Product name: RY200S: Fine silica (manufactured by Nippon Aerosil), specific surface area 76m by BET method ² / G, primary particle diameter (average value) 16 nm, silica purity 99.9% or more (indicating the purity of silica as the main inorganic component in the whole inorganic component), apparent specific gravity 50 g / l, water content 0.089 PH value of mass%, carbon content 4.3 mass%, 4% dispersion (water: methanol = 1: 1 solution) 5.72
[0043]
(2) Hydrophobization treatment of core substrate
As the substrate (2), a copper-clad substrate made of BT (bismaleimide triazine) resin having a thickness of 800 μm is used. After drilling by mechanical drilling, a conductor layer (4) is formed by copper plating with a thickness of 15 μm on the inner wall surface (3) of the hole to produce a core substrate (1) having a through hole with a diameter of 300 μm (FIG. 2). The surface of the conductor layer of the obtained substrate is roughened using a commercially available micro-etching solution (trade name: MEC etch BOND, manufactured by MEC) to form a roughened surface (5) (FIG. 3). See). Then, a rust preventive treatment (trade name: CZ treatment, manufactured by MEC) is performed with a commercially available rust preventive agent to form a hydrophobic surface (6), and the hydrophobic treatment is completed (see FIG. 4).
[0044]
When the contact angle (2θ) with respect to water of the surface of the conductor layer subjected to the hydrophobization treatment was measured by a liquid method using a contact angle measuring device (trade name: CA-A, manufactured by Kyowa Kagaku), the contact angle (2θ) was 101. It was a degree. The contact angle (2θ) here refers to the imaginary line (82) drawn from the contact point (80) with the conductor surface (7) of the water drop (8) to the apex (81) of the water drop and the horizontal line (83). This is the double angle (2θ) of the angle (θ) (9) to be formed (see FIG. 5).
[0045]
(3) Cavity test
Each through-hole filling paste shown in Table 1 is filled into the through-holes of the core substrate (1). Non-woven paper is placed on a pedestal with a vacuum suction device, and the substrate is placed on the pedestal. On top of that, a stainless steel filling mask having three types of through holes of 200 μm, 300 μm, and 400 μm is installed so as to correspond to the positions of the through holes. Next, each hole filling paste in Table 1 is placed, and hole filling is performed while pressing a roller squeegee. Then, after temporarily curing on the condition of 120 ° C. × 40 minutes, the composition is cured on the condition of 150 ° C. × 5 hours and thermally cured. Here, the hole filling property is evaluated by setting the surface of the cured through-hole filler (11) to have a recess (d) (10) exceeding 20 μm as a failure (see FIG. 6). Those having a pass rate of 85% or more in 100 through holes are judged as pass. Pass / fail is also shown in Table 1.
[0046]
▲ 4 ▼ Gap observation
The through-hole portion of the core substrate is cut, and the presence or absence of a gap between the cured through-hole fillers is observed using a magnifying microscope with a magnification of 200 times. Those having a pass rate of 85% or more in 100 through holes are judged as pass. Pass / fail is also shown in Table 1.
[0047]
[Table 1]

[0048]
In Sample No. 1 to Sample No. 3, which are examples using an imidazole-based curing agent having the structure of the general formula (1), no gap is generated, and good results are obtained. In particular, R ₁ In Sample No. 1 and Sample No. 2 in which is a methyl group, almost no dent was generated (10 μm or less), and good hole filling was achieved.
[0049]
On the other hand, in Sample No. 4 and Sample No. 5, which are comparative examples using an imidazole curing agent having no hydroxyl group in the molecule, a gap was generated inside. Sample No. 6 as a reference example was obtained by changing the copper filler of sample No. 4 as a comparative example to a silica filler, but no gap was generated. From this, it can be seen that this gap is an inherent phenomenon when a copper filler is used.
[0050]
Example 2 Example of FC-PGA type multilayer printed wiring board
(1) Fabrication of semiconductor device using FC-PGA type multilayer printed wiring board
A multilayer printed wiring board using the method for manufacturing a printed wiring board of the present invention will be described below using an example of a so-called “FC-PGA” structure shown in FIG.
As shown in FIG. 8, a FR-5 double-sided copper-clad core substrate in which a copper foil (200) having a thickness of 18 μm is bonded to an insulating substrate (101) having a thickness of 0.4 mm is prepared. The core substrate used here has a TMA Tg (glass transition point) of 175 ° C., a substrate surface direction CTE (thermal expansion coefficient) of 16 ppm / ° C., and a substrate surface vertical direction CTE (thermal expansion coefficient) of 50 ppm / The dielectric constant ε at 4.7 ° C. and 1 MHz is tan δ at 1 MHz is 0.018.
[0051]
A photoresist film is attached on the core substrate, and exposure and development are performed to provide an opening having a diameter of 600 μm and an opening (not shown) corresponding to a predetermined wiring shape. The copper foil exposed at the opening of the photoresist film is removed by etching using an etching solution containing sodium sulfite and sulfuric acid. The photoresist film is peeled and removed to obtain a core substrate on which an exposed portion (300) as shown in FIG. 9 and an exposed portion (not shown) corresponding to a predetermined wiring shape are formed.
[0052]
An etching process is performed by a commercially available etching processing apparatus (CZ processing apparatus manufactured by MEC) to roughen the surface of the copper foil, and then an insulating film having a thickness of 35 μm mainly composed of an epoxy resin is attached to both surfaces of the core substrate. And it cures on the conditions of 170 degreeC x 1.5 hours, and forms an insulating layer. The properties of the insulating layer after curing are as follows: Tg (glass transition point) by TMA is 155 ° C., Tg (glass transition point) by DMA is 204 ° C., CTE (thermal expansion coefficient) is 66 ppm / ° C., dielectric constant ε at 1 MHz 3.7, tan δ at 1 MHz is 0.033, weight loss at 300 ° C. is −0.1%, water absorption is 0.8%, moisture absorption is 1%, Young's modulus is 3 GHz, tensile strength is 63 MPa, The elongation is 4.6%.
[0053]
As shown in FIG. 10, via holes (500) for interlayer connection are formed in the insulating layer (400) using a carbon dioxide laser. The form of the via hole is a slot shape with a surface layer portion having a diameter of 120 μm and a bottom portion having a diameter of 60 μm. Further, the output of the carbon dioxide gas laser is increased to form a through hole (600) so as to penetrate the insulating layer and the core substrate. There are three types of through-holes (not shown) having a diameter of 200 μm, 300 μm, and 400 μm. The inner wall surfaces of these through holes have undulations (not shown) unique to laser processing. And after immersing a board | substrate in the catalyst activation liquid containing palladium chloride, electroless copper plating is given to the whole surface (not shown).
[0054]
Next, copper panel plating (700) having a thickness of 18 μm is applied to the entire surface of the substrate. Here, a via hole conductor (800) for electrically connecting the layers is formed in the via hole. A through-hole conductor (900) that electrically connects the front and back surfaces of the substrate is formed in the through-hole. The diameter of the through hole after panel plating is 160 μm, 260 μm, and 360 μm. Etching is performed by a commercially available etching processing apparatus (CZ processing apparatus manufactured by MEC) to roughen the surface of the copper plating. Thereafter, a rust preventive treatment (trade name: CZ treatment) is applied with the company's rust preventive agent to form a hydrophobic surface, thereby completing the hydrophobic treatment. When the contact angle (2θ) with respect to water of the surface of the conductor layer subjected to the hydrophobization treatment was measured by a liquid method using a contact angle measuring device (trade name: CA-A, manufactured by Kyowa Kagaku), the contact angle (2θ) was 101. It was a degree.
[0055]
Non-woven paper is placed on a pedestal with a vacuum suction device, and the substrate is placed on the pedestal (not shown). A stainless steel hole filling mask having three types of through holes of 200 μm, 300 μm, and 400 μm is installed thereon (not shown) so as to correspond to the positions of the through holes. Next, each through-hole filling paste shown in Table 1 is placed, and filling is performed while filling a roller squeegee (not shown).
[0056]
As shown in FIG. 11, the through-hole filling paste (102) filled in the through-hole is temporarily cured under the condition of 120 ° C. × 20 minutes. Next, as shown in FIG. 12, after polishing the surface of the core substrate using a belt sander (rough polishing), the surface is flattened by buffing (final polishing) and cured under conditions of 150 ° C. × 5 hours. Complete the hole filling process. A part of the substrate that has completed this hole filling step is used for the hole filling property evaluation test.
[0057]
As shown in FIG. 13, a 8 mm through hole (110) is formed using a mold (not shown). As shown in FIG. 14, a masking tape (120) is attached to one surface of the substrate. And as shown in FIG. 15, eight multilayer chip capacitors (130) are arrange | positioned using the chip mounter on the adhesion surface of the masking tape (120) exposed to the through-hole (110). This multilayer chip capacitor is composed of a laminate (150) in which insulators and conductors are alternately laminated, and an electrode (140) protrudes 70 μm from the laminate.
[0058]
As shown in FIG. 16, an embedded resin (160) mainly composed of an epoxy resin and a silica filler is filled into the through-hole in which the multilayer chip capacitor is disposed using a dispenser (not shown). The embedded resin is defoamed and thermally cured under the conditions of a primary heating step of 80 ° C. × 3 hours and a secondary heating step of 170 ° C. × 6 hours.
[0059]
As shown in FIG. 17, the surface of the cured embedded resin is roughly polished using a belt sander, and then finish-polished by lapping. The end face of the electrode of the chip capacitor is exposed on the polished surface. Next, the temporarily cured hole filling resin is cured under conditions of 150 ° C. × 5 hours.
[0060]
Then, swelling liquid and KMnO _Four Using the solution, the polishing surface of the embedded resin is roughened. After activating the Pd catalyst on the roughened surface, copper plating is performed in the order of electroless plating and electrolytic plating. As shown in FIG. 18, the plating layer formed on the embedded resin is electrically connected to the end face of the electrode of the chip capacitor. A resist is formed on the plated surface, and a predetermined wiring pattern is patterned. Unnecessary copper is replaced with Na ₂ S ₂ O ₈ Etching away using concentrated sulfuric acid. The resist is removed to complete the formation of the wiring as shown in FIG. Etching is performed with a commercially available etching processing apparatus (CZ processing apparatus manufactured by MEC) to roughen the surface of the copper plating of the wiring.
[0061]
A film (190) serving as an insulating layer is laminated thereon and thermally cured, and then irradiated with a carbon dioxide laser to form via holes for interlayer connection. The surface of the insulating layer is roughened using the same oxidizing agent as described above, and a predetermined wiring (200) is formed by the same method. A dry film to be a solder resist layer is laminated on the outermost surface of the wiring board, and the mounting pattern of the semiconductor element is formed by exposing and developing to complete the formation of the solder resist layer (210). A predetermined wiring (230) and a solder resist layer (240) are formed on the surface to be pinned for mounting by a similar method to obtain a multilayer printed wiring board before pinning as shown in FIG. .
[0062]
The terminal electrode (200) for mounting the semiconductor element is plated in the order of Ni plating and Au plating (not shown). After solder paste made of low melting point solder is printed thereon, a solder bump (220) for mounting a semiconductor element is formed through a solder reflow furnace.
[0063]
On the other hand, on the opposite side of the semiconductor element mounting surface, after solder paste made of high melting point solder is printed, solder bumps (260) for pinning through a solder reflow furnace are formed. In a state where the pins (250) are set on a jig (not shown) and the substrate is placed, pins are attached through a solder reflow furnace (not shown). As shown in FIG. An FC-PGA type multilayer printed wiring board before mounting is obtained. When the amount of positional deviation from the predetermined position of the tip of the pin was measured using a projector, a good result of 0.1 mm or less was obtained.
[0064]
The semiconductor element (270) is disposed on the semiconductor element mounting surface at a position where it can be mounted, and the semiconductor element is mounted through a solder reflow furnace under a temperature condition where only the low melting point solder is melted. After the mounting portion is filled with an underfill material with a dispenser, it is thermally cured to obtain a semiconductor device using an FC-PGA type multilayer printed wiring board on which a semiconductor element as shown in FIG. 7 is mounted.
[0065]
(2) Evaluation of through-hole fillability
A hole filling property evaluation test is performed using the substrate that has completed the hole filling step. Using a magnifying glass with a magnification of 200 times from the end surface of the through hole filler, the presence or absence of defects such as cracks is visually inspected.
[0066]
On the other hand, in the comparative example, the hole filling process is a known screen printing method using a metal mask, and is performed under the conditions shown in Table 2. The diameter of the through hole of the metal mask is the same as that of the support plate of the embodiment. After completion of the filling, the through-hole filler is visually inspected for defects using a magnifying glass having a magnification of 200 times from the end surface of the through-hole. The results are shown in Table 2. Moreover, the presence or absence of defects, such as a crack, is visually inspected from a through-hole end surface of the cured through-hole filler using a magnifying glass having a magnification of 200 times.
[0067]
[Table 2]

[0068]
From the results of Table 2, according to the present invention, even if it is a through-hole formed so as to penetrate the substrate having the build-up layer formed on both sides of the core substrate and having a undulation on the inner wall surface, It can be seen that filling and curing are possible without causing defects such as gaps during curing. Even when there are through holes having different diameters in one substrate, the filling property can be made uniform. Further, since unevenness and gaps are not generated in the vicinity of the through hole opening end face, unevenness is not generated in the subsequently formed buildup layer. Therefore, it is suitable for a multilayer printed wiring board that is pinned to a buildup layer immediately below the through hole, such as an FC-PGA type wiring board.
[0069]
Example 3: Evaluation of the effect of ultrafine inorganic particles on the through-hole filling paste
(1) Adjustment of paste for filling through holes
Except that the characteristic values are different as described in Table 3, the composition is adjusted to the paste for filling through-holes using the composition of Sample No. 1 of Example 1. The adjustment method is also performed in accordance with Example 1. Some samples use alumina and titania instead of aerosil.
(2) Evaluation
A hole-filling test is performed using the same method as in Example 1. The results are: ◎ (substantially 100% pass) for those with extremely good hole filling properties, ○ (95% to less than 100% pass) for those with considerably good hole filling properties, and Δ (85% or more for those with good hole filling properties). The results are shown in Table 3. In addition, the change with time of the viscosity is obtained by leaving the paste for through hole filling for 2 months. The state after being left standing is shown in Table 3 together with “◎” indicating almost no increase in viscosity, “◯” indicating only a slight increase in viscosity, and “Δ” indicating an increase in viscosity.
[0070]
[Table 3]

[0071]
From the results, it can be seen that by adjusting the characteristics of the ultrafine inorganic filler to a specific range, not only the hole filling property but also the change with time of the viscosity can be effectively suppressed. The critical significance of the characteristic value of the ultrafine inorganic filler will be explained below.
[0072]
From Sample No. 13, it can be seen that if the specific surface area is larger than this, the viscosity becomes too high and the hole filling property is not good. It can also be seen that the lower limit of the primary particle diameter is at most 5 nm. From Sample No. 28, it can be seen that if the specific surface area becomes smaller than this, the change with time in viscosity is not good. Looking at sample number 26, it can be seen that if the primary particle size is larger than this, the change in viscosity over time is not necessarily large.
[0073]
From Sample No. 14, it can be seen that if the pH value is further reduced, the change in viscosity over time is not good. The details of this reason are unknown, but when it becomes acidic, it is presumed that an undesirable side reaction has occurred with the basic curing agent. Also, from sample No. 19, it can be seen that if the pH value is further increased, the change in viscosity with time is not necessarily increased. Although the details of this reason are unknown, it is presumed that some undesired side reaction occurs when it is more basic than necessary.
[0074]
Looking at Sample No. 21, it can be seen that if the carbon content is further reduced, the hole filling performance is lowered. It is presumed that the carbon component suppresses aggregation due to intermolecular force or the like and suppresses fluctuations in viscosity. Sample No. 20 having a carbon content of 5% by mass does not noticeably deteriorate the characteristics as in Sample No. 21, but if it is included more than this, it is not preferable because it interferes with plating on the through hole.
[0075]
Looking at Sample No. 22, it can be seen that if the water content is further increased, the change in viscosity over time is not good. It is presumed that moisture has affected the curability of the resin. Moreover, when the sample number 23 is seen, even if a water content decreases more than this, it turns out that a time-dependent change of a viscosity may not increase a little. Probably, it is considered that the humidity of the working environment is easily absorbed.
[0076]
【The invention's effect】
According to the present invention, in a through-hole filler containing a base metal filler such as copper, a gap is generated inside the through-hole filler after curing without causing a decrease in heat resistance and occurrence of dents on the front and back surfaces. Can be prevented. Further, since unevenness and gaps are not generated in the vicinity of the through hole opening end face, unevenness is not generated in the subsequently formed buildup layer. Since a substrate with excellent flatness of the build-up layer in the vicinity of the through hole can be obtained, a PGA (pin grid array) type pinned to the build-up layer immediately below the through hole like an FC-PGA type wiring substrate. This is particularly suitable when forming a multilayer printed wiring board.
[Brief description of the drawings]
FIG. 1 is an explanatory view showing a gap generated in a through-hole filler after curing.
FIG. 2 is an explanatory view showing a process of hydrophobizing the core substrate.
FIG. 3 is an explanatory view showing a process of hydrophobizing the core substrate.
FIG. 4 is an explanatory view showing a process of hydrophobizing the core substrate.
FIG. 5 is an explanatory view showing a method of measuring a contact angle θ with respect to water on a conductor surface.
FIG. 6 is an explanatory view showing a dent d generated on the surface of the through-hole filler after curing.
FIG. 7 is an explanatory diagram of a semiconductor device using an FC-PGA type multilayer printed wiring board used in an example of the present invention.
FIG. 8 is a schematic view of a copper-clad core substrate having a thickness of 400 μm used in an example of the present invention.
FIG. 9 is an explanatory view showing a state after patterning of a copper-clad core substrate having a thickness of 400 μm.
FIG. 10 is an explanatory view showing a state in which via holes and through holes are formed in a substrate in which an insulating layer is formed on both surfaces of a core substrate.
FIG. 11 is an explanatory view showing a state after panel plating is applied to a substrate having an insulating layer formed on both surfaces of a core substrate.
FIG. 12 is an explanatory diagram of a substrate in which through holes are filled and filled.
FIG. 13 is an explanatory view showing a substrate in which through holes are formed by punching.
FIG. 14 is an explanatory view showing a state in which a masking tape is attached to one surface of a substrate in which through holes are formed by punching.
FIG. 15 is an explanatory view showing a state in which a multilayer chip capacitor is arranged on a masking tape exposed in a through hole.
FIG. 16 is an explanatory view showing a state in which an embedded resin is filled in a through hole.
FIG. 17 is an explanatory view showing a state where a substrate surface is polished and flattened.
FIG. 18 is an explanatory diagram showing a state in which panel plating is applied to a polished surface of a substrate.
FIG. 19 is an explanatory diagram showing a state in which wiring has been hatched.
FIG. 20 is an explanatory view showing a state in which a buildup layer and a solder resist layer are formed on a substrate.
FIG. 21 is an explanatory diagram of an FC-PGA type multilayer printed wiring board used in an example of the present invention.
[Explanation of symbols]
1 Core substrate
2 Substrate
The inner wall surface of 3 holes
4 Conductor layer
5 Roughened surface
6 Hydrophobic surface
7 Conductor layer surface
8 water drops
80 Contact point of water drop with conductor layer surface
81 The top of a water drop
82 Virtual lines
83 horizontal lines
9 Angle (θ)
10 Cured through-hole filler
11 dent (d)
12 Clearance

Claims (9)

An epoxy resin, a curing agent, and a metal filler contains an inorganic filler, a through hole filling paste without phenol resin, the metal filler is a powder comprising at least a base metal, curing agent or less imidazole compounds der represented by the general formula (1) in is, the inorganic filler is Mu through-hole filling a specific surface area by the BET method, wherein ultrafine inorganic filler der Rukoto of 40 to 400 2 / ^g Solvent paste.
General formula (1)

(In the formula, R ₁ represents a hydrogen atom, an alkyl group having 1 to 10 carbon atoms, a hydroxyalkyl group having 1 to 10 carbon atoms, or an alkyloxy group having 1 to 10 carbon atoms.) ^{2. The} solventless paste for filling through holes according to claim 1, wherein the ultrafine inorganic filler has a specific surface area of 60 to 100 m ² / g by BET method . An epoxy resin, a curing agent, a metal filler, an inorganic filler, and a through-hole filling paste that does not contain a phenol resin, wherein the metal filler is a powder containing at least a base metal, and the curing agent is as follows: A solventless paste for filling through-holes, wherein the inorganic filler is an ultrafine inorganic filler having a primary particle size of 5 to 50 nm .
General formula (1)

(In the formula, R ₁ represents a hydrogen atom, an alkyl group having 1 to 10 carbon atoms, a hydroxyalkyl group having 1 to 10 carbon atoms, or an alkyloxy group having 1 to 10 carbon atoms.) The solventless paste for filling through holes according to claim 3, wherein the ultrafine inorganic filler has a primary particle diameter of 10 to 20 nm . An epoxy resin, a curing agent, a metal filler, an inorganic filler, and a paste for through-hole filling that does not contain a phenol resin, the metal filler is a powder containing at least a base metal, and the curing agent is: A solvent-free paste for filling through-holes , characterized in that the inorganic filler is an ultrafine inorganic filler having an apparent specific gravity of 30 to 200 g / l .
General formula (1)

(In the formula, R ₁ represents a hydrogen atom, an alkyl group having 1 to 10 carbon atoms, a hydroxyalkyl group having 1 to 10 carbon atoms, or an alkyloxy group having 1 to 10 carbon atoms.) 6. The ultrafine inorganic filler having a pH value of 3.5 to 6.5 in a 4% dispersion (water: methanol = 1: 1 solution) is used. Solvent-free paste for filling through-holes according to any one of the above. The through-hole filling according to any one of claims 1 to 6, wherein the ultrafine inorganic filler used has a carbon content of 3 to 5% by mass when the organic component is converted to carbon . Solvent-free paste. The solventless paste for filling through-holes according to any one of claims 1 to 7, wherein the ultrafine inorganic filler has a water content of 0.05 to 0.15 mass% . The through-hole filling according to any one of claims 1 to 8, wherein the ultrafine inorganic filler has a purity of 99.5% or more as a main inorganic component in the whole inorganic component excluding an organic component. Solvent-free paste for use.

Images