The Study of Anisotropic Copper Etching with

博士 === 國立清華大學 === 化學工程學系 === 91 === Copper etching becomes a difficult process for outer layers of printed circuit boards (PCBs) because of the circuitry miniaturization. Ammoniacal cupric solution is one of the most common etchants used because of its high etching rate (40~70m min-1). However, it...

Full description

Bibliographic Details
Main Authors: Chao-Wen, Shih, 史朝文
Other Authors: Chi-Chao, Wan
Format: Others
Language:en_US
Published: 2003
Online Access:http://ndltd.ncl.edu.tw/handle/54597111804082633008
id ndltd-TW-091NTHU0063070
record_format oai_dc
collection NDLTD
language en_US
format Others
sources NDLTD
description 博士 === 國立清華大學 === 化學工程學系 === 91 === Copper etching becomes a difficult process for outer layers of printed circuit boards (PCBs) because of the circuitry miniaturization. Ammoniacal cupric solution is one of the most common etchants used because of its high etching rate (40~70m min-1). However, it has serious undercut problem. In addition, the toxic odor of ammoniacal etchant also induces need for substitution. In this dissertation, monoethanolamine (MEA, H2NCH2CH2OH), which is less volatile, was used to replace ammonia as the chelating agent of cupric ion for preparing the etchant to fabricate fine lines. Electrochemical polarization methods were used to investigate the dissolution mechanism and the effects of the chelating agents and additives. The dissolution mechanism of copper in this MEA-complexed cupric solution (CuSO4 + MEA) in the presence of bridging ligands like F-, Cl-, Br-,I-, SCN- was first studied. The shift of the corrosion potential (Ecorr) as opposed to the original solution corresponds to increased copper dissolution rate due to the ligands. So we propose that copper dissolution proceeds through an “inner-sphere” pathway in solution containing bridging ligands. In contrast, some conventional oxidizers including S2O82-, Cr2O72-, MnO4- were also added in the etchant and the Ecorr did not shift obviously. The order of effectiveness of these ligands is as follows, SCN- > I- > Br- > Cl- > F-, which is related to their polarizability. MEA-complexed cupric solution containing 1M CuCl2 and 3.3 to 10M MEA was then studied at various temperatures (25 to 55℃) and pH values (10 to 6.5). The effects of these factors on dissolution rate and etch factor of the copper patterns of PCBs were also discussed. It was found that the highest corrosion current density (icorr) was obtained with [MEA] being around 5M and the decreasing icorr in response to increasing [MAE] indicating the inhibition behavior of MEA. Activation energies (Ea) of MEA-complexed cupric solutions were measured and the heat of adsorption (ΔHads), which accounts for the chemisorption of the MEA ligands on the copper surface, was calculated. ΔHads was found to increase with solution containing excess MEA ([MEA] > 5M), again indicating the inhibition behavior of MEA. In addition, at lower pH, icorr increased because the concentration of MEA ligands decreased due to the acid reaction. Further, high etch factor can be achieved on PCBs with 75m/75m, line/space (L/S) with etchants containing high MEA concentration, which means the inhibitive property of MEA renders obvious banking effect. The inhibitive property of MEA in copper dissolution was further investigated. It has been found that in oxygen sparging condition, the dissolution rate increases as the concentration of MEA increases, however, when the concentration of MEA exceeds 1.25M, the dissolution rate decreases. The inhibitive property in concentrated MEA solution is attributed to the competitive adsorption between MEA and oxygen. When the pH of the MEA solution is adjusted to more acidic condition, MEA molecule will be protonated and can not adsorb on copper. Furthermore, the increasing intensity of nitrogen and suppression of copper from data by the ESCA analysis indicates the formation of Cu : MEA. Moreover, a comparative study of copper etching using various amine-complexed cupric chloride solutions was carried out. Solutions containing ammonia, ethyl-, monoethanol-, diethanol-, triethanol- and ethylenediamine and oxygen without cupric chloride have shown that the reaction occurs in two regions, dissolution region at low concentration ( < 0.2M) and inhibition region at higher concentration ( > 0.2M). It is considered that the tail functional group of the amine affects the dissolution rate. In addition, the interaction between copper and amine-complexed ion like [Cu(II)(NH3)4]2+, [Cu(II)(NH2CH2CH2)4]2+ [Cu(II)(MEA)4]2+, and [Cu(II)(DEA)4]2+ can induce copper dissolution. However, [Cu(II)(En)2]2+ and [Cu(II)(TEA)]2+ can cause inhibition for copper dissolution. Steric effect of the ligands seems to be the cause. In addition, chloride ion was added to increase the dissolution rate. The electron is transferred from the copper surface onto the cupric species through the bridging ligand, which greatly influences the copper dissolution rate. However, the effect is not seen in the case of [Cu(II)(En)2]2+ and [Cu(II)(TEA)]2+. Finally, copper pattern etching was tested on PCB with 50m/50m, line/space (L/S) by spray copper etching with these non-ammonia etchants. Improved etching factor can be achieved with non-ammoniacal etchants including [Cu(II)(MEA)4Cl]+ and [Cu(II)(DEA)4Cl]+. It means the tail functional group of the amine with its inhibitive property can reduce the undercut. Although the etching rate of non-ammonia-complexed cupric etchant is still lower than the ammonia-complexed etchant, the etching factor of the former is significantly higher.
author2 Chi-Chao, Wan
author_facet Chi-Chao, Wan
Chao-Wen, Shih
史朝文
author Chao-Wen, Shih
史朝文
spellingShingle Chao-Wen, Shih
史朝文
The Study of Anisotropic Copper Etching with
author_sort Chao-Wen, Shih
title The Study of Anisotropic Copper Etching with
title_short The Study of Anisotropic Copper Etching with
title_full The Study of Anisotropic Copper Etching with
title_fullStr The Study of Anisotropic Copper Etching with
title_full_unstemmed The Study of Anisotropic Copper Etching with
title_sort study of anisotropic copper etching with
publishDate 2003
url http://ndltd.ncl.edu.tw/handle/54597111804082633008
work_keys_str_mv AT chaowenshih thestudyofanisotropiccopperetchingwith
AT shǐcháowén thestudyofanisotropiccopperetchingwith
AT chaowenshih fēiānxìshítóngyèzàifēiděngxiàngxìngshíkèzhīyánjiū
AT shǐcháowén fēiānxìshítóngyèzàifēiděngxiàngxìngshíkèzhīyánjiū
AT chaowenshih studyofanisotropiccopperetchingwith
AT shǐcháowén studyofanisotropiccopperetchingwith
_version_ 1718319012305174528
spelling ndltd-TW-091NTHU00630702016-06-22T04:21:07Z http://ndltd.ncl.edu.tw/handle/54597111804082633008 The Study of Anisotropic Copper Etching with 非氨系蝕銅液在非等向性蝕刻之研究 Chao-Wen, Shih 史朝文 博士 國立清華大學 化學工程學系 91 Copper etching becomes a difficult process for outer layers of printed circuit boards (PCBs) because of the circuitry miniaturization. Ammoniacal cupric solution is one of the most common etchants used because of its high etching rate (40~70m min-1). However, it has serious undercut problem. In addition, the toxic odor of ammoniacal etchant also induces need for substitution. In this dissertation, monoethanolamine (MEA, H2NCH2CH2OH), which is less volatile, was used to replace ammonia as the chelating agent of cupric ion for preparing the etchant to fabricate fine lines. Electrochemical polarization methods were used to investigate the dissolution mechanism and the effects of the chelating agents and additives. The dissolution mechanism of copper in this MEA-complexed cupric solution (CuSO4 + MEA) in the presence of bridging ligands like F-, Cl-, Br-,I-, SCN- was first studied. The shift of the corrosion potential (Ecorr) as opposed to the original solution corresponds to increased copper dissolution rate due to the ligands. So we propose that copper dissolution proceeds through an “inner-sphere” pathway in solution containing bridging ligands. In contrast, some conventional oxidizers including S2O82-, Cr2O72-, MnO4- were also added in the etchant and the Ecorr did not shift obviously. The order of effectiveness of these ligands is as follows, SCN- > I- > Br- > Cl- > F-, which is related to their polarizability. MEA-complexed cupric solution containing 1M CuCl2 and 3.3 to 10M MEA was then studied at various temperatures (25 to 55℃) and pH values (10 to 6.5). The effects of these factors on dissolution rate and etch factor of the copper patterns of PCBs were also discussed. It was found that the highest corrosion current density (icorr) was obtained with [MEA] being around 5M and the decreasing icorr in response to increasing [MAE] indicating the inhibition behavior of MEA. Activation energies (Ea) of MEA-complexed cupric solutions were measured and the heat of adsorption (ΔHads), which accounts for the chemisorption of the MEA ligands on the copper surface, was calculated. ΔHads was found to increase with solution containing excess MEA ([MEA] > 5M), again indicating the inhibition behavior of MEA. In addition, at lower pH, icorr increased because the concentration of MEA ligands decreased due to the acid reaction. Further, high etch factor can be achieved on PCBs with 75m/75m, line/space (L/S) with etchants containing high MEA concentration, which means the inhibitive property of MEA renders obvious banking effect. The inhibitive property of MEA in copper dissolution was further investigated. It has been found that in oxygen sparging condition, the dissolution rate increases as the concentration of MEA increases, however, when the concentration of MEA exceeds 1.25M, the dissolution rate decreases. The inhibitive property in concentrated MEA solution is attributed to the competitive adsorption between MEA and oxygen. When the pH of the MEA solution is adjusted to more acidic condition, MEA molecule will be protonated and can not adsorb on copper. Furthermore, the increasing intensity of nitrogen and suppression of copper from data by the ESCA analysis indicates the formation of Cu : MEA. Moreover, a comparative study of copper etching using various amine-complexed cupric chloride solutions was carried out. Solutions containing ammonia, ethyl-, monoethanol-, diethanol-, triethanol- and ethylenediamine and oxygen without cupric chloride have shown that the reaction occurs in two regions, dissolution region at low concentration ( < 0.2M) and inhibition region at higher concentration ( > 0.2M). It is considered that the tail functional group of the amine affects the dissolution rate. In addition, the interaction between copper and amine-complexed ion like [Cu(II)(NH3)4]2+, [Cu(II)(NH2CH2CH2)4]2+ [Cu(II)(MEA)4]2+, and [Cu(II)(DEA)4]2+ can induce copper dissolution. However, [Cu(II)(En)2]2+ and [Cu(II)(TEA)]2+ can cause inhibition for copper dissolution. Steric effect of the ligands seems to be the cause. In addition, chloride ion was added to increase the dissolution rate. The electron is transferred from the copper surface onto the cupric species through the bridging ligand, which greatly influences the copper dissolution rate. However, the effect is not seen in the case of [Cu(II)(En)2]2+ and [Cu(II)(TEA)]2+. Finally, copper pattern etching was tested on PCB with 50m/50m, line/space (L/S) by spray copper etching with these non-ammonia etchants. Improved etching factor can be achieved with non-ammoniacal etchants including [Cu(II)(MEA)4Cl]+ and [Cu(II)(DEA)4Cl]+. It means the tail functional group of the amine with its inhibitive property can reduce the undercut. Although the etching rate of non-ammonia-complexed cupric etchant is still lower than the ammonia-complexed etchant, the etching factor of the former is significantly higher. Chi-Chao, Wan Yun-Yung, Wang 萬其超 王詠雲 2003 學位論文 ; thesis 160 en_US