U.S. patent application number 14/508385 was filed with the patent office on 2015-11-26 for aluminum alloy composite and method for joining thereof.
This patent application is currently assigned to TAISEI PLAS CO., LTD.. The applicant listed for this patent is TAISEI PLAS CO., LTD.. Invention is credited to Naoki Andoh, Masanori Naritomi.
Application Number | 20150336355 14/508385 |
Document ID | / |
Family ID | 39765777 |
Filed Date | 2015-11-26 |
United States Patent
Application |
20150336355 |
Kind Code |
A1 |
Naritomi; Masanori ; et
al. |
November 26, 2015 |
ALUMINUM ALLOY COMPOSITE AND METHOD FOR JOINING THEREOF
Abstract
The object of the present invention is to strongly join an
aluminum alloy part with an FRP prepreg. An object obtained by
subjecting an aluminum alloy to a suitable liquid treatment so as
to form a surface having large, micron-order irregularities and
also fine irregularities with a period of several tens of
nanometers, eliminating the presence of sodium ions from the
surface and additionally forming a surface film of aluminum oxide,
which is thicker than a natural oxide layer, has been found to have
a powerful adhesive strength with epoxy-based adhesives. By
simultaneously curing an FRP prepreg which uses the same
epoxy-based adhesive in the matrix, an integral composite or
structure in which FRP and aluminum alloy have been united at a
joining strength of unprecedented magnitude is produced.
Inventors: |
Naritomi; Masanori; (Tokyo,
JP) ; Andoh; Naoki; (Tokyo, JP) |
|
Applicant: |
Name |
City |
State |
Country |
Type |
TAISEI PLAS CO., LTD. |
Tokyo |
|
JP |
|
|
Assignee: |
TAISEI PLAS CO., LTD.
Tokyo
JP
|
Family ID: |
39765777 |
Appl. No.: |
14/508385 |
Filed: |
October 7, 2014 |
Related U.S. Patent Documents
|
|
|
|
|
|
Application
Number |
Filing Date |
Patent Number |
|
|
12530940 |
Sep 11, 2009 |
8932719 |
|
|
PCT/JP2008/054539 |
Mar 12, 2008 |
|
|
|
14508385 |
|
|
|
|
Current U.S.
Class: |
428/137 ;
428/336 |
Current CPC
Class: |
B32B 27/38 20130101;
B32B 2262/106 20130101; B32B 2255/26 20130101; B32B 2264/102
20130101; H05K 3/386 20130101; B32B 27/04 20130101; B32B 2260/021
20130101; Y10T 428/31529 20150401; B32B 2307/306 20130101; H05K
3/385 20130101; C09J 163/00 20130101; B32B 2457/00 20130101; B32B
2264/101 20130101; B32B 2307/54 20130101; B32B 2255/06 20130101;
B32B 5/147 20130101; B32B 5/08 20130101; B32B 2262/101 20130101;
B32B 38/08 20130101; Y10T 428/265 20150115; B32B 2605/18 20130101;
Y10T 428/24322 20150115; Y10T 428/24355 20150115; H05K 3/383
20130101; B32B 2605/00 20130101; B32B 15/08 20130101; B32B
2262/0269 20130101; B32B 2260/046 20130101; B32B 2307/714 20130101;
B32B 5/024 20130101; B32B 2264/107 20130101; H05K 1/056 20130101;
B32B 7/12 20130101; C23C 22/83 20130101; B32B 2264/104 20130101;
Y10T 428/31522 20150401; H05K 2203/0315 20130101; B32B 15/20
20130101; B32B 15/14 20130101; C23F 1/36 20130101; B32B 43/00
20130101; B32B 2307/542 20130101 |
International
Class: |
B32B 15/08 20060101
B32B015/08; B32B 27/38 20060101 B32B027/38; B32B 15/20 20060101
B32B015/20; B32B 7/12 20060101 B32B007/12; B32B 15/14 20060101
B32B015/14 |
Foreign Application Data
Date |
Code |
Application Number |
Mar 12, 2007 |
JP |
2007-062376 |
Claims
1. An aluminum alloy composite comprising: a first metal part made
of a first aluminum alloy having a micron-order surface roughness
formed by chemical etching on a surface thereof, the surface having
thereon a thin layer of aluminum oxide with a thickness of at least
2 nm that is free of sodium ions and the surface roughness having
formed therein an ultrafine irregular surface formed of depressions
or projections having 10 to 100 nm diameter and substantially equal
depth or height, and an adherent part joined to the first metal
part with an epoxy-based adhesive that has penetrated into the
ultrafine irregular surface; wherein said micron-order surface
roughness has a mean peak spacing: RSm of 0.8 to 10 .mu.m and a
maximum height: Rz of 0.2 to 5.0 .mu.m; wherein said adherent part
is fiber-reinforced plastic which contains the epoxy-based adhesive
and is filled, laminated and reinforced with at least one selected
from among continuous fibers, short fibers and fiber cloth, wherein
said epoxy-based adhesive in the cured form includes a total of
between 0 and 100 parts by weight of filler per 100 parts by weight
of the total resin components, and wherein the shear strength at
break with said first metal part and said second metal part is 44
to 60 MPa.
2. The aluminum alloy composite according to claim 1, wherein the
shear strength at break with said first metal part and said
fiber-reinforced plastic as the adherent part is 50 to 60 MPa.
3. The aluminum alloy composite of claim 1, which is obtained by
joining a plate part made of the aluminum alloy and a plate object
of the fiber-reinforced plastic and has formed therein a
throughhole for passing a bolt or rivet therethrough.
Description
CROSS-REFERENCE TO RELATED APPLICATIONS
[0001] This application is a division of application Ser. No.
12/530,940 filed on Sep. 11, 2009, which is a national stage of
international application no. PCT/JP2008/054539 filed on Mar. 12,
2008, which claims foreign priority of Japanese Patent Application
2007-062376 filed on Mar. 12, 2007 in Japan.
FIELD OF THE INVENTION
[0002] The present invention relates to methods for joining metals
with metals or metals with plastics that are used in transportation
machinery, electrical equipments, medical instruments, general
machinery, other industrial machinery, consumer equipments or the
like; to composites thereof; and to methods for manufacturing the
same. More specifically, the present invention relates to aluminum
alloy composites and joining methods for integrally joining optimal
aluminum alloy parts and plastic parts together, suitable for use
as transportation machinery components that are required to be
lightweight, such as automotive parts and aircraft parts.
BACKGROUND OF THE INVENTION
[0003] With the recent rise of energy prices, further technical
innovations are being sought in aircraft. That is, in the new model
aircraft and the new model concepts recently announced by the
Boeing Company (U.S.) and Airbus (France), the utilization of
aluminum alloys such as extra super duralumin (designated as
"A7075" according to the Japanese Industrial Standards (JIS)) and
extra duralumin (similarly designated as "A2024") for reducing the
weight of the fuselage has been abruptly decreased and the
utilization of carbon fiber reinforced plastic (abbreviated below
as "CFRP") is being increased by a corresponding amount.
[0004] A7075 (extra super duralumin), which is one type of aluminum
alloy, has a specific gravity of about 2.7, whereas CFRP has a
specific gravity of 1.6 to 1.7, making it incomparably lighter. Up
until now, the utilization of CFRP as a material for aircraft has
been increased in military applications such as fighter planes and
combat helicopters, while use as a civilian aircraft material has
not grown so much as anticipated due to the need to develop
know-how on manufacturing large components and because of obstacles
from the standpoint of cost. The use of CFRP has been avoided in
civilian applications, in spite of their light weight, high
strength and high corrosion resistance. Lately, however, in
addition to research and development by the above two leading
civilian aircraft manufacturers, the rise of the price of crude oil
and its duration have made it imperative to examine ways of
reducing the airframe weight.
[0005] The automotive industry as well, in order to cope with the
rise of crude oil prices and environmental issues, is promoting the
development of hybrid vehicles, electric cars and, in Europe,
high-performance diesel engine vehicles. Moreover, it is expected
that fuel cell vehicles, which have the highest energy conversion
efficiency, will someday see widespread use. Although significant
advances have been made in the development of fuel cells, the
largest and most difficult problem that remains for use in
automobiles is how to safely and smoothly handle hydrogen fuel, for
which laborious work has been made. While safety is most important
in civilian use, the obstacles are too large to make it a routine
matter for general users to deal with cars loaded with hydrogen gas
under several hundred atmospheres of pressure. Hence, the current
liquid fuel system is not likely to change for the time being. In
this sense, reducing the weight of vehicle construction is
important for achieving real improvements in the power system. In
fact, aluminum alloy materials are being applied increasingly in
some types of vehicles and it is predicted that use of CFRP will be
required someday.
[0006] In aircraft manufacture, the increased utilization of CFRP
materials entails an increase in material expenses due to the
high-cost CFRP materials, in addition to which it presents
technical problems. One such problems is that, even in the latest
Airbus and Boeing aircraft for which mass production has already
been decided, it is estimated that the utilization of aluminum
alloys will top 50% in terms of weight, with both CFRP materials
and A7075 materials being used. CFRP materials will be used in
areas close to the wing tips and A7075 materials will be used, as
in the past, in central structure areas of the fuselage. Joining of
the two materials will generally be done with rivets or with bolts
and nuts, while special products will be required for this
purpose.
[0007] This is because of the large differences between CFRP
materials and metals in their basic properties. Metal materials
have a large elongation (elongation at break), that of A7075 being
10 to 16%, whereas CFRP has an elongation of only several percent.
When a strong tensile force is applied to these materials, the
metal material undergoes elastic stretch (in accordance with
Young's modulus, elongation and contraction that is proportional to
the force) up to a certain strength; when a force that exceeds this
limit is applied, elongation in excess of Young's modulus takes
place, where break occurs at 110 to 116% of the original length
(100%) for A7075. On the other hand, for CFRP, when pulled in a
direction parallel to the fibers, elongation of the carbon fibers
themselves is only 1 to 2%. When a large tensile load is applied
that exceeds the range of elongation and contraction in accordance
with Young's modulus, the carbon fibers break and the CFRP tears.
In other words, CFRP has a small range in the ability of the
material to absorb forces by undergoing elongation itself.
[0008] The same is true not only for tensile forces but also for
crushing forces (compressive forces). That is, when tightened with
bolts and nuts, even if the compressive forces exceed a certain
limit, a metal will be deformed itself, enabling break to be
avoided. However, when a compressive load is applied to CFRP, the
force is first supported by the cured epoxy resin; with the
application of an excessive pressing force, the epoxy resin is
forced to spread peripherally and be deformed but spreading is
limited by the carbon fibers, as a result of which the resin does
not move and thus cannot be deformed, leading ultimately to
failure.
[0009] In short, when a throughhole is formed in CFRP, a bolt is
inserted into the hole and a nut threaded on the bolt is used to
tighten the CFRP with an excessive torque, the CFRP undergoes
compressive break. Because the large differences between the
physical properties of both materials are due to inherent
differences between a metal part, which is composed of metallic
bonds between atoms, and epoxy resins and carbon fibers, which are
composed of covalent bonds between carbon atoms and the like, there
are basically no means for improving the properties themselves.
Therefore, when both are tightened and secured with bolts and nuts,
the only solution is to avoid applying excess pressure so as to
prevent break on the CFRP side. This requires the development and
use of special bolts and nuts. It has even been reported that the
success by a certain corporation in developing such a bolt
structure helped to spur the competition today in development of
civilian aircrafts.
[0010] In future aircrafts, regardless of the degree to which the
utilization of CFRP increases, the utilization of light metal
materials will never go to zero, which means that techniques for
easily joining CFRP materials with aluminum alloy materials will
continue to be very important basic technology. There is another
problem as to releasing of the CFRP prepreg from the metallic mold
after its heat-curing. The prepreg undergoes heat-curing in a
pressurized state under compressive forces applied by the metallic
mold. In this operation, because the epoxy resin functions as an
adhesive with respect to the mold as well, it is necessary to apply
a release agent between the mold and the prepreg. As a result,
infiltration of the release agent (generally a silicone oil-type
release agent) into the CFRP product is unavoidable, which makes it
impossible to achieve the highest physical properties inherent to
the epoxy resin. Even minor decreases in quality are issues that
must be resolved for use in structures for high-speed moving
machinery such as aircrafts and automobiles. The present invention
was conceived in order to provide a solution to such problems.
[0011] The inventors have made inventions as to techniques for
securely joining a plastic part formed by injection molding with a
metal part that has been inserted beforehand into the metallic mold
for injection, specifically, aluminum alloy parts, magnesium alloy
parts, copper alloy parts, titanium alloy parts, stainless steel
parts, etc. (which techniques are referred to below as "injection
joining" techniques) (see Patent Document 1: WO 03/064150 A1,
Patent Document 2: WO 2004/041532 A1, Patent Document 4: Japanese
Patent Application No. 2006-329410, Patent Document 5: Japanese
Patent Application No. 2006-281961, Patent Document 6: Japanese
Patent Application No. 2006-345273) and Patent Document 6: Japanese
Patent Application No. 2006-354636).
[0012] A major factor was the discovery of surface treatment
methods carried out beforehand on the metal to be inserted. The
inventors anticipated that the surface shape of the metal obtained
by such surface treatment would have a desirable effect not only on
injection joining but also on joining (adhesion) with ordinary
adhesives.
[0013] In the above-described inventions as to "injection joining",
the surface state desired in the metal alloy to be used can be
summarized in terms of the following conditions (1) to (3). The
first of these conditions (1) is that the rough surface obtained by
chemical etching have irregularities (depressions and protrusions)
with a period of 1 to 10 .mu.m, with the height difference between
the depressions and protrusions being about one-half of the period,
i.e., 0.5 to 5 .mu.m. The reason for this is that, when the molten
resin flows into the metallic mold having a temperature well over a
hundred and several decade degrees Centigrade lower than the
melting point of the resin under a high pressure of several hundred
to a thousand atmospheres, the diameter of the depressions into
which the resin can somehow manage to penetrate as it crystallizes
and hardens is 1 to 10 .mu.m.
[0014] However, it is actually difficult that cover 100% of an
aluminum alloy surface in this way with such a rough surface, given
the variability of chemical reactions. It is practically regarded
that such a surface roughness satisfies the above-mentioned
roughness conditions, for which a curve indicating depressions and
protrusions with an irregular period ranging from 0.2 to 20 .mu.m
and having a maximum height difference in a range of 0.2 to 10
.mu.m as measured with a surface roughness tester can be traced, or
for which a mean peak spacing (RSm) is 0.8 to 10 .mu.m and a
maximum height (Rz) is 0.2 to 10 .mu.m analyzed with a scanning
probe microscope, where such amounts as RSm and Rz are defined in
the JIS standard (JIS B 0601:2001). The inventors, having
recognized and concluded that the period of the depressions and
protrusions on an ideal rough surface is, as noted above, 1 to 10
.mu.m, refer in the present invention to a rough surface with a
surface roughness defined in this way as a "surface with
micron-order roughness," which is a readily understood technical
term.
[0015] The second condition (2) is that, when the rough surface is
viewed with a magnification of an electron microscopic level, it
has a finely irregular surface of depressions and protrusions with
a period of 10 to 500 nm, most preferably a finely irregular
surface with a period of 40 to 50 nm. The third condition (3) is
that the surface is covered with a thin layer of metal oxide which
is either thicker or stronger than an ordinary natural oxide layer
for that particular metal alloy. Each of these three conditions
required on the metal alloy side were attained, as noted above, for
all of the following: magnesium alloys, titanium alloys, copper
alloys, stainless steel alloys, aluminum alloys, etc., as a result
of which it was possible to achieve a high shear breaking strength
between the metal and the cured resin of at least 20 to 30 MPa with
injection joining. This clearly proved the correctness of the
hypothesis that the above three conditions are essential for
injection joining. At this point, the inventors expected that this
hypothesis should obviously have desirable effects also in joining
(adhesion) with adhesives.
[0016] Hence, the inventors proposed the following hypothesis
concerning adhesive joining. First, the metal alloy piece with a
surface which satisfies the three above mentioned conditions is
prepared in the same way as that used in the above described
injection joining experiment and a liquid, one-pack epoxy adhesive
is applied to the metal piece. Next, the metal piece is subjected
to such steps as being placed in a vacuum state and then returned
to standard pressure, causing the adhesive to infiltrate and
smoothly coat the finely textured face of the metal alloy surface.
The adhesive is then cured through heating. It was thought that, in
this way, the liquid epoxy adhesive would be able to enter into the
depressions of the micron-order roughness of condition (1) for the
metal alloy surface at an inflow pressure of only about one
atmosphere. If such penetration is possible, the epoxy adhesive
will be cured within these valleys with subsequent heating.
[0017] In this case, the inner walls of these depressions have the
nanometer-level finely irregular surface of condition (2). Complete
penetration of the epoxy adhesive to the depths of the fine
depressions formed in the finely irregular surface of (2) is
presumed to be difficult. However, some of the epoxy adhesive does
extend inside the openings of the fine depressions and is cured. In
such a case, the epoxy adhesive that has been cured inside a large
depression achieves the state of being secured (engaged) at the
interior of the depression by innumerable spikes, which should make
it difficult to be peeled away from the metal substrate by an
external force. When the cured epoxy resin is forcibly peeled away,
because the metal oxide layer of above-described condition (3) for
the surface in contact with the spikes has a sufficient thickness
to exhibit the hardness of this ceramic material, deformation on
the metal alloy side is limited, making it impossible to pull out
the epoxy resin within the large depression. Ultimately, the epoxy
resin itself breaks near the opening of the large depression. In
such cases, the force required for break far exceeds the adhesive
strength data for hitherto known adhesives.
[0018] The correctness of this hypothesis has, in fact, already
been proven, first for aluminum alloys, then for magnesium alloys,
copper alloys, titanium alloys and stainless steel alloys.
Subsequent to the present invention, the inventors are occupied in
verification tests and plan to propose a group of inventions
concerning various metal alloys. Based on these tests, the
inventors believe that the above hypothesis concerning adhesive
joining is correct, while it will accept the approval or criticism
from many scientists and chemists. In the present invention, the
inventors refer to this hypothesis as "NAT (Nano Adhesion
Technology)". With NAT, as explained above in regards to adhesion
with an adhesive, joining can be understood as an entirely physical
effect, that is, as an anchoring effect. Without such an
understanding, it is impossible to explain the immense joining
strengths of 500 to 700 kgf/cm.sup.2 (50 to 70 N/mm.sup.2=50 to 70
MPa), measured as the shear strength at break and the tensile
strength at break, which are exhibited with the use of an
epoxy-based adhesive, not only for aluminum alloys but also for
other metal alloys as well.
[0019] In addition, it has been mentioned above as condition (1)
that large depressions and protrusions, that is, depressions and
protrusions with a period of 1 to 10 .mu.m, are desirable, while
this NAT applies not only to the aluminum alloys shown in the
present invention but it has been demonstrated also, for example,
for copper alloys, titanium alloys, stainless steel alloys and
common steel materials. In cases where depressions and protrusions
larger than described above are present and, conversely, in cases
where depressions and protrusions that are too much smaller than
described above are present, the joining strength due to joining
with an adhesive became lower. For depressions that are too large,
the most likely reason is such that the density of depressions that
forms per unit surface area becomes low, thus reducing the
anchoring effect. For depressions that are too small, penetration
of the epoxy adhesive into the interior is probably
insufficient.
[0020] By utilizing the above-described joining strength according
to NAT, it is possible to achieve the many needs mentioned in the
preceding section. First, in the case of adhesive joining between
metal alloys for which surface treatment according to NAT
hypothesis was made, a very strong adhesive force can be obtained
between metal alloys by using an epoxy adhesive. This is the case
both in joining between aluminum alloys and in joining between an
aluminum alloy and a titanium alloy. The reason is that the joining
strength itself does not arise between the metals but it rather
arises between the respective metals and the epoxy resin. Also, a
fiber-reinforced plastic (FRP) material in which the epoxy resin is
used as the matrix is, not surprisingly, the most trouble-free
mating material for adhesive joining with a metal alloy piece. It
can be appreciated that, by pressing together a FRP prepreg and an
aluminum alloy part that has been coated with an epoxy adhesive,
raising the temperature and simultaneously curing the epoxy resins
on both sides, joining (anchoring) will be easier than between
metals.
[0021] One conceivable form of this adhesion is a structure in
which the FRP material is sandwiched between thin sheets of
aluminum alloy, that is, a laminated structure. Although there is
an increase in weight, degradation of the epoxy resin by a release
agent can be prevented, because a mold release agent for the
prepreg is not necessary. Also, in cases where a sandwich structure
is formed in which part of the prepreg rather than the entire
surface is placed between thick sheets of aluminum alloy to form a
sandwich structure, throughholes are formed in the structure, bolts
are passed through the holes and joining to another part is carried
out, break of the CFRP portion can be avoided even when the bolts
are over-tightened.
[0022] Additionally, in a plate-like or tubular structural member
formed integrally of an aluminum alloy member at the ends and a
CFRP member as the main material at the center, connection with
bolts and nuts, fitting or some other known metal joining method
may be employed by utilizing the ends thereof, facilitating
assembly and disassembly, thus giving a member compatible with
large-volume production. This should be helpful in reducing weight
and increasing strength not only in aircrafts but also in vehicles
such as automobiles, in mobile electronic and electrical
equipments, in robot machineries or the like. Today, CFRP materials
are very familiar materials. The ability to employ such materials
in vehicle applications will enable major contributions to be made
to the energy-saving and environmentally responsible society in the
future.
SUMMARY OF THE INVENTION
[0023] The present invention, which was conceived in order to solve
the problems described above, achieves the following objects.
[0024] One object of the present invention is to joining an
aluminum alloy composite obtained by joining a CFRP material with
an aluminum alloy and a method for manufacturing such a
composite.
[0025] A further object of the present invention is to provide an
aluminum alloy composite and a method for manufacturing the same,
which enable a CFRP material and an aluminum alloy part to be
joined without using a fastening means such as a bolt and nut
having a special construction, that is, which enables such joining
to be carried out with ordinary fastening means.
[0026] A still further object of the invention is to provide an
aluminum alloy composite and a method for manufacturing the same,
which do not require a release agent when a CFRP material and an
aluminum alloy are joined (anchored).
[0027] The invention employs the following means to achieve the
above objects.
[0028] The aluminum alloy composite of the present invention is
composed essentially of a first metal part made of a first aluminum
alloy having a micron-order surface roughness formed by chemical
etching on a surface thereof, the surface having thereon a thin
layer of aluminum oxide with a thickness of at least 2 nm that is
free of sodium ions and the surface roughness having formed therein
an ultra-fine irregular surface formed of depressions or
projections having 10 to 100 nm diameter and substantially equal
depth or height; and
[0029] an adherent part joined to the first metal part with an
epoxy-based adhesive that has penetrated into the ultrafine
irregular surface.
[0030] The method for manufacturing an aluminum alloy composite of
the present invention is composed essentially of a step of shaping
an aluminum alloy part by mechanical working from a casting or an
intermediate material; a chemical etching step by immersing said
shaped aluminum alloy part in a strongly basic aqueous solution; a
neutralizing step by immersing said aluminum alloy part in an
aqueous solution of an acid after the chemical etching step; an
ultrafine etching step by immersing the aluminum alloy part in an
aqueous solution containing at least one selected from among
hydrazine hydrate, ammonia and water-soluble amine compounds so as
to form on the part an ultrafine irregular surface after the
neutralizing step; a step of coating said ultrafine irregular
surface of the aluminum alloy part with an epoxy adhesive; a step
of forming a fiber-reinforced plastic prepreg material containing
epoxy resin component into a required shape; a step of attaching
the shaped prepreg material to the epoxy adhesive-coated surface of
the aluminum alloy part; and a step of positioning said prepreg
material and the aluminum alloy part and curing the epoxy-based
adhesive and the epoxy resin component through applying heat and
pressure.
BRIEF DESCRIPTION OF THE DRAWINGS
[0031] FIG. 1 is a schematic view showing a firing jig for curing
an aluminum alloy piece and a CFRP prepreg within a hot-air dryer
and a method of setting up the same.
[0032] FIG. 2 shows a composite, which is also a test specimen for
measuring the joining strength therebetween by applying a tensile
load to break, obtained by joining an aluminum alloy sheet and a
CFRP prepreg with an epoxy-based adhesive.
[0033] FIG. 3 is a schematic view showing an exemplified
construction obtained by using bolts and nuts to join an aluminum
alloy piece-FRP prepreg composite with another metal structural
member.
[0034] FIG. 4 shows a test specimen obtained by joining together
two pieces of aluminum alloy with an epoxy adhesive, the purpose of
the specimen being to measure the joining strength between the
aluminum alloy pieces by applying a tensile load to break.
[0035] FIG. 5 is an electron micrograph of a piece of A7075
aluminum alloy that was etched with an aqueous solution of caustic
soda.
[0036] FIG. 6 is an electron micrograph of a piece of A7075
aluminum alloy that was etched in an aqueous solution of caustic
soda and then neutralized in an aqueous solution of nitric
acid.
[0037] FIG. 7 is an electron micrograph of a piece of A7075
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid and then
etched in an aqueous solution of hydrazine hydrate as an ultrafine
etching.
[0038] FIG. 8 is an electron micrograph of a piece of A7075
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid, then
etched in an aqueous solution of hydrazine hydrate as an ultrafine
etching and subsequently treated with aqueous hydrogen
peroxide.
[0039] FIG. 9 is an electron micrograph of a piece of A5052
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid and then
subjected to ultrafine etching treatment for 1 minute in an aqueous
solution of hydrazine hydrate having a concentration of 3.5%.
[0040] FIG. 10 is an electron micrograph of a piece of A5052
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid and then
subjected to ultrafine etching treatment for 2 minutes in an
aqueous solution of hydrazine hydrate having a concentration of
3.5%.
[0041] FIG. 11 is an electron micrograph of a piece of A5052
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid and then
subjected to ultrafine etching treatment for 4 minutes in an
aqueous solution of hydrazine hydrate having a concentration of
3.5%.
[0042] FIG. 12 is an electron micrograph of a piece of A5052
aluminum alloy that was etched with an aqueous solution of caustic
soda, neutralized with an aqueous solution of nitric acid and then
subjected to ultrafine etching treatment for 8 minutes in an
aqueous solution of hydrazine hydrate having a concentration of
3.5%.
[0043] FIG. 13 is a graph of the results presented in Table 1.
[0044] FIG. 14 is a graph of the results presented in Table 2.
[0045] FIG. 15 shows the results by observation of the A5052
aluminum alloy obtained in Experiment example 1 with a scanning
probe microscope.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0046] The constituent features of these inventions will be
described respectively below in detail.
[Aluminum Alloy Part]
[0047] The aluminum alloy used in the present invention may be of
any type, so long as it is an aluminum alloy. For example, all
malleable aluminum alloys, such as those defined in Japanese
Industrial Standard (JIS) A1000 series to 7000 series (e.g.,
corrosion-resistant aluminum alloys, high-strength aluminum alloys,
heat-resistant aluminum alloys, etc.) and aluminum alloys for
casting, such as JIS ADC 1 to 12 (aluminum alloys for die casting),
may be used. In the case of casting alloys, shaped forms that may
be used include parts that have been shaped by dye casting or such
parts that have additionally been shaped by mechanical working. In
the case of malleable alloys, a plate stock or the like as an
intermediate material and a part, which has been shaped from the
same by mechanical working such as hot pressing, may be used.
[Surface Treatment, Pre-Treatment of Aluminum Alloy Parts: Theories
and Basic Thought Thereof]
[0048] It is preferable to immerse the aluminum alloy part at first
in a degreasing tank and remove any fats, oils and oil-based agents
that have adhered to the part in mechanical working. A special
degreasing treatment is not required in the present invention.
Rather, it is desirable to prepare a warm aqueous solution by
adding a commercially available degreasing agent for aluminum
alloys to hot water in the concentration specified by the chemical
manufacturer and rinse the aluminum alloy part by immersing it in
the solution. That is, degreasing treatment by a method commonly
used on aluminum alloys will suffice. Although there will be some
variation depending on the particular degreaser product, with an
ordinary commercially available product, immersion is carried out
for a period of 5 to 10 minutes at a concentration of 5 to 10% and
at a solution temperature of 50 to 80.degree. C.
[0049] In subsequent pretreatment steps, the treatment method will
differ according to whether the aluminum alloy includes a
relatively large amount of silicon or has only a low level of such
ingredients. In the case of alloys having a low silicon content,
i.e., malleable aluminum alloys such as A1050, A1100, A2014, A2024,
A3003, A5052 and A7075, a treatment method like the following is
preferred. That is, it is desirable for promoting good
reproducibility of the subsequent alkali etching step to immerse
the aluminum alloy part in an acidic aqueous solution for a short
time and then to rinse with water, thus allowing the acid
ingredient to be adsorbed onto the surface layer of the aluminum
alloy part. This treatment, which could be referred to as a
preliminary pickling step, may be carried out using a dilute
aqueous solution of an inexpensive mineral acid such as nitric
acid, hydrochloric acid or sulfuric acid at a concentration of one
to several percent. Next, etching is carried out by immersing the
aluminum alloy part in a strongly basic aqueous solution and
rinsing with water.
[0050] Such etching strips away fats, oils and scum remaining on
the surface of the aluminum alloy along with the surface layer of
the aluminum alloy. Simultaneously with such stripping, the surface
comes to have a micron-level roughness, i.e., a mean peak spacing
(RSm) of 0.8 to 10 .mu.m and a maximum height (Rz) of 0.2 to 5.0
.mu.m in terms of the JIS standards (JIS B 0601: 2001; ISO 4287:
1997; ISO 1302: 2002). When examined with a scanning probe
microscope of the type in use today, these numerical values are
automatically calculated and output of them is given. However,
there are cases where, when numerical values are displayed for the
fine surface irregularities by automated output, the calculated RSm
values do not represent the actual state. For better results, it is
necessary to visually reconfirm the RSm values on a roughness curve
graph that can be generated as output by the scanning probe
microscope and represent this irregular surface state.
[0051] If, on visual inspection of the roughness curve graph, the
roughness state is one where the height difference falls in a range
of 0.2 to 5 .mu.m at irregular periods within a range of 0.2 to 20
.mu.m, the actual state is substantially the same as that mentioned
above. Such visual inspection method is advantageous because, in
cases where automatic calculations are judged to be unreliable, a
determination can easily be made by visual inspection. To
summarize, using the technical term defined herein, the surface of
the aluminum alloy part is formed to be "a surface with
micron-order roughness." The solution used is preferably an aqueous
solution of caustic soda at a concentration of one to several
percent, with immersion being carried out for several minutes at
preferably 30 to 40.degree. C. Next, it is preferable to complete
pretreatment by again immersing the aluminum alloy part in an
acidic aqueous solution, then rinsing with water to remove sodium
ions. The inventors refer to this as the neutralization step. It is
especially preferable for this acidic aqueous solution to be an
aqueous solution of nitric acid having a concentration of several
percent.
[0052] In the case of aluminum alloys for casting, such as ADC 10
and ADC 12, it is preferable for the procedure to include the
following steps. That is, following the degreasing step in which
oils and fats are removed from the surface of the aluminum alloy,
it is desirable to carry out preliminary pickling and etching in
the same way as in the above-described step. As a result of such
etching, copper components or silicon components which do not
dissolve under strongly basic conditions become fine-grained black
smut (because such contaminants are referred to in the plating
industry as smut, this usage is followed here as well). It is thus
desirable to dissolve and strip away this smut, then immerse the
aluminum alloy part in an aqueous solution of nitric acid having a
concentration of several percent. With immersion in an aqueous
solution of nitric acid, the copper smut is dissolved and the
silicon smut rises off the aluminum alloy surface.
[0053] In the case where the alloy used is one having a large
amount of silicon components, such as ADC 12, if the alloy is only
immersed in an aqueous solution of nitric acid, silicon smut will
continue to adhere to the surface of the aluminum alloy substrate
and will not be completely stripped away. Therefore, it is
preferable to carry out ultrasonic cleaning in which the part is
immersed in a water tank and ultrasonic vibration is given to the
water, thereby physically stripping off the silicon smut. While
this does not remove all the smut, it is sufficient for practical
purposes. Pretreatment may be brought to an end at this point,
although it is preferable to once again immerse the part in a
dilute aqueous solution of nitric acid for a short time and rinse
it with water. This completes pretreatment. Because pretreatment
ends with immersion in an acidic aqueous solution and rinsing with
water, no sodium ions remain. The sodium ions are explained
below.
[0054] Based on experimental findings, the joining strength which
is given when two pieces of aluminum alloy are joined with each
other using an epoxy adhesive is determined substantially by the
micron-order roughness and the nanometer-order, ultrafine irregular
shape characteristics of the surfaces with depressions and
protrusions. The experimental findings show that, by searching for
the immersion conditions, etc. when etching is carried out with an
aqueous solution of caustic soda, if such conditions mentioned
above that the NAT hypothesis are satisfied in respect of surface
configuration, even if it be occasional, a surprisingly strong
adhesive strength can be obtained. However, once surface treatment
has been finished in case of treatment that involves only etching
with caustic soda, sodium ions will remain in the aluminum alloy
surface layer even with more than adequate water rinsing
thereafter. Because sodium ions have a small particle size, they
migrate easily. Even after painting or joining has been carried
out, in a wet state of the overall assembly, sodium ions which have
remained present are entrained by water molecules passing through
the resin layer and are collected at the metal/resin interfacial
surface for some reason, where they promote oxidation of the
aluminum surface.
[0055] That is, corrosion of the aluminum alloy surface occurs, as
a result of which separation between the substrate and the paint
film or adhesive is promoted. Given such circumstances, there is as
yet no reason to carry out etching in an aqueous caustic soda
solution as an aluminum alloy pretreatment prior to joining.
Accordingly, even today, the standard pretreatment method for
strong adhesive joining is regarded to be chromate treatment, in
which the aluminum alloy is immersed in an aqueous hexavalent
chromium compound solution of potassium bichromate and chromium
trioxide, or anodization and use in an unsealed state. In short,
prior to pay attention to enhancing the adhesive strength by
etching, primary aim was taken at prevention of corrosion and
degradation of the aluminum alloy surface.
[0056] However, the etching of aluminum alloys with caustic sodium
has not been abandoned altogether and is still commonly used in
pretreatment for painting. Generally, an extreme adhesive strength
is not required in painting, for it is thought that, other than in
outdoor use applications where the product is placed in the
weather, immersion in water will not occur. Moreover, unless the
product is of a type where, for example, the paint film is to be
guaranteed for ten years, such painting pretreatment is not
unreasonable. Without being based on such easy-going reasoning, the
main subject of the present invention is to achieve a long-term
joining stability. To this end, the most important issue is to get
rid of sodium ions.
Sodium in Aluminum Alloy
[0057] The sodium (Na) present in the aluminum alloy will be also
discussed. In the aluminum metal manufacturing process, bauxite is
dissolved in an aqueous solution of caustic soda to give a
high-purity aluminum compound, which is then electrolytically
reduced to yield aluminum metal. In this manufacturing process, the
presence of sodium as an impurity in the aluminum metal is
unavoidable. However, in current metallurgical technology, it is
possible to suppress the sodium content in aluminum alloys to an
extreme degree. Therefore, in an ordinary environment in which
there is no acid-base mist, corrosion does not advance in
commercially available aluminum alloys today when direct wetting
(liquid water) is not present. Corrosion actually proceeds at a
high rate when there is wetting, salt (sodium chloride) from sea
winds and heating with sunlight. That is, when commercially
available aluminum alloys are used outdoors in coarse environments,
such as in towns and cities located near the seashore where sea
winds are strong and the atmospheric temperature is high, the rate
of corrosion is high.
[0058] The solution to avoid such corrosion is generally to coat
the entire surface with paint, adhesive or the like. When this is
done, it is essential that cracks or fissures are not formed in the
paint film or adhesive layer and further it is important to keep
salt-containing water from entering such cracks and fissures and
penetrating to the surface of the aluminum alloy. When such
measures are taken, it is not always necessary for surface
treatment of the aluminum alloy to entail common chromate
treatment. As long as the paint film has a good weather resistance
and adhesion between the paint film and the substrate is good, even
in an adverse environment, the aluminum alloy will be durable long
enough with painting alone. In particular, with the use of
hexavalent chromium currently being in the course of worldwide
rejection, chromate treatment can no longer be regarded as a
desirable method for surface treatment of aluminum alloys. Today,
many paints with excellent durability and adhesives having
outstanding moisture and heat resistance are commercially
available. In light of the above, the inventors have attempted to
optimize the conditions to be sought on the aluminum alloy side, in
order to maintain a strong joining between paint or adhesive and
the aluminum alloy substrate for a long time and to arrive at a
theoretical understanding of such conditions.
Surface of the Aluminum Alloy
[0059] The desired degree of roughness for the aluminum alloy
surface is obtained primarily with a strongly basic aqueous
solution such as a caustic soda solution, followed by immersion in
an aqueous solution of acid and thorough rinsing with water to
remove sodium ions. However, under observation with an electron
microscope, the microstructure of the surface obtained by etching
with an aqueous solution of caustic soda (FIG. 5 shows an example
of such a surface on A7075) has very fine depressions and
protrusions with a periodicity of several tens of nanometers. Such
a surface should make a cured adhesive difficult to be removed from
depressions in the substrate; indeed, a finely irregular surface
such as this is sought after and desirable according to the NAT
hypothesis. Yet, on the surface obtained by immersing this aluminum
alloy in an aqueous solution of nitric acid and rinsing with water
(FIG. 6 shows an example of such a surface on A7075), the quality
required for the fine depressions and protrusions was degraded. In
short, the operation of immersing the aluminum alloy in an acidic
aqueous solution to remove sodium ions amounted to a type of
chemical polishing. The surface, as viewed in an electron
micrograph, forms what might be expressed in sensory terms as a
textured surface. The degree of texture on this textured surface is
lowered by chemical polishing, which has an opposite effect on
adhesive joining.
[0060] Accordingly, the approach taken was to restore this degree
of texture by the main treatment described below. Micrographs for
reference are shown in FIGS. 5 to 7, with FIG. 6 being an example
of an electron micrograph taken after immersion treatment in an
aqueous solution of nitric acid. The reasons explained above will
be appreciated from these images. It should be noted that the
experiences, thought and theories by the inventors which ultimately
led to the present invention are also a product of the ability to
make ready use of high-performance electron microscopes capable of
achieving a high resolution of several nanometers. Also, the
thought in the present invention was to achieve durability and
corrosion resistance of the aluminum alloy by having the final
aluminum alloy surface be surface layer of aluminum oxide and by
raising the joining strength of the adhesive to the alloy substrate
to the utmost.
[Surface Treatment, Main Treatment and Ultrafine Etching of the
Aluminum Alloy Part]
[0061] The aluminum alloy part on which pretreatment has been
completed is then subjected to the following type of surface
treatment as the final treatment, i.e., main treatment. Following
the completion of pretreatment, the aluminum alloy part is immersed
in an aqueous solution containing at least one from among hydrazine
hydrate, ammonia and a water-soluble amine compound, after which it
is preferably rinsed with water and dried at 70.degree. C. or
below. This is also intended to restore a rough surface after the
surface was somewhat altered by the sodium ion-removing treatment
carried out as the final step in pretreatment (so that it became
smoother while retaining a degree of roughness). The aluminum alloy
part is immersed for a short time in a weakly basic aqueous
solution, such as an aqueous solution of hydrazine hydrate, thereby
making ultrafine etching carried out, causing the surface to be
covered with a ultrafine irregular surface having depressions or
protrusions of 10 to 100 nm diameter and equal height or depth.
More precisely, it is desirable to form numerous fine
irregularities having a period of 40 to 50 nm over the inner walls
of the depressions in the micron-order irregularities so as to
finish the surface to a high degree of what might be referred to as
texture when viewed on an electron micrograph.
[0062] Also, by making the drying temperature after rinsing
elevated to at least 100.degree. C., if the interior of the dryer
is hermetically sealed, hydroxylation reactions between boiling
water and aluminum arise, altering the surface and forming a
boehmite layer. This can hardly be called a tough surface layer and
is thus undesirable. The moisture conditions within the dryer are
determined not only by the size of the dryer and the manner of air
circulation but also by the amount of aluminum alloy loaded
therein. In this sense, regardless of the loading conditions,
hot-air drying at 90.degree. C. or lower, preferably at 70.degree.
C. or lower, is desirable for achieving good results having a good
reproducibility to prevent the formation of boehmite at the
surface. When drying is carried out at below 70.degree. C., only
trivalent aluminum can be detected from the aluminum peaks in
surface elemental analysis by XPS; the zero-valent aluminum which
can be detected in XPS analysis in commercially available A5052 and
A7075 aluminum alloy plate stock and the like vanishes.
[0063] Because XPS analysis allows elements present from the metal
surface to a depth of 1 to 2 nm to be detected, these results show
that immersion in an aqueous solution of hydrazine hydrate or an
amine compound followed by rinsing with water and hot-air drying
have the effect of making the natural oxide layer (a thin aluminum
oxide layer having a thickness of about 1 nm) initially present on
the aluminum alloy even thicker through the main treatment. It was
found that, unlike the natural oxide layer, the layer has a
thickness of at least 2 nm, so further study was not performed.
That is, although analysis at deeper positions of from about 10 to
100 nm is possible by carrying out XPS analysis after etching with
an argon ion beam or the like, it was concluded that such analysis
would be difficult at the present time, because there is a
possibility that the valence number of aluminum atoms in deeper
layers will change due to the influence of the beam itself. As a
result, the inventors suspended such inquiry.
[0064] The formation of an aluminum oxide layer on the aluminum
alloy surface by other surface treatment methods will be discussed.
One surface treatment method carried out to improve the durability
of aluminum alloy is anodization. When anodization has been carried
out on an aluminum alloy, an aluminum oxide layer having a
thickness of from several microns (.mu.m) to somewhat over ten
microns (.mu.m) can be formed, greatly enhancing durability.
Innumerable openings in pits having a diameter of about 20 to 40 nm
remain on the aluminum oxide layer immediately after anodizing
treatment. If adhesive joining or paint application is carried out
in this state, that is, in an unsealed alumite state, the adhesive
or paint will enter somewhat through the openings into the pits and
be cured, apparently exhibiting a strong anchoring effect and
giving rise to a high joining strength in joining with an adhesive.
As a matter of fact, the joining of a dissimilar material with such
an anodized aluminum alloy by coating the aluminum alloy with an
adhesive is known in aircraft assembly.
[0065] However, the inventors have doubts about this. That is, when
shear breaking tests were carried out on a united object obtained
by strongly joining together two pieces of anodized aluminum alloy
using an epoxy adhesive, no sample broke with forces in excess of
40 MPa (40 N/mm.sup.2), according to the breaking tests by the
inventors. Moreover, on examining the faces of fracture, it was not
the adhesive that broke but rather, in most cases, the anodized
layer (aluminum oxide layer) was separated from the aluminum alloy
substrate. The inventors concluded from this that, "the surface of
the metal side required for strong joining must be a ceramic-type
high-hardness layer such as a metal oxide, while it must not be too
thick." Although the anodic oxide surface layer is aluminum oxide,
which is an oxide of the substrate aluminum itself, the surface
layer is a ceramic and the substrate is a metal, so the types of
material are different each other.
[0066] If the ceramic material is thick, differences in the
physical properties will necessarily emerge in an extreme state, as
a result of which the material is likely to break. For this reason,
it is preferable for the metal oxide layer to be thin. Also, it is
generally understood that if the metal oxide is a ceramic in an
amorphous or microcrystalline state, it should be joined firmly
with the substrate. That is, it was concluded that, in order to
bring the shear breaking strength and tensile breaking strength of
the adhesive to a very high level of 50 to 100 MPa, the metal oxide
layer should not be made so thick and that using unsealed alumite
which has been anodized is undesirable.
Ultrafine Etching
[0067] What is referred to as ultrafine etching in the present
invention will be now described in further details. Upon immersion
of the aluminum alloy in a weakly basic aqueous solution having a
pH of 9 to 10, e.g., hydrazine hydrate, ammonia or a water-soluble
amine, at a suitable temperature and for a suitable length of time,
the entire surface becomes covered with ultrafine irregular shapes
with depressions and protrusions having a diameter of 10 to 100 nm.
Expressed as a number-average diameter, this is about 50 nm. Stated
differently, this means that it is desirable to select a pH,
temperature and time which are optimal for obtaining ultrafine
topographically irregular shapes having a diameter of 10 to 100 nm
on the surface. Based on experience, the inventors predict that the
ultrafine irregularities with depressions and protrusions have a
period or a diameter, which is most preferably about 50 nm. The
reason is that, in the case of irregularities having a period of 10
nm, rather than being a textured surface, the manner of
irregularity is too fine, resulting in a surface that is smooth for
an adhesive. On the other hand, a period of more than 100 nm
appears to exceed the bounds of what would normally be construed as
a textured surface, so that there is little inclination to
anchoring. In the present invention, "number average" does not
refer to a mean of summation of a degree that enables statistical
validation but, rather, it refers to an average value obtained for
20 or fewer samples.
[0068] The value of 50 nm was obtained at empirically based on
experimental results. However, even when aiming at a period of 50
nm, a perfectly ordered arrangement cannot be obtained with
chemical reactions, so some variability is to be expected. There is
no alternative but to examine photographs taken with an electron
microscope and convert the resultant into numerical values. The
result corresponds to a surface of ultrafine irregular shapes that
are substantially entirely (100%) covered with 10 to 100 nm
diameter depressions and protrusions of the same depth or height.
In fact, when most of the surface was occupied by depressions or
protrusions having a diameter of 10 to 20 nm, or, conversely, by
depressions and protrusions having a diameter of 100 nm or more,
the joining strength was inferior. Actual cases, in which A7075
material or A5052 material was etched with an aqueous solution of
hydrazine hydrate, will be described below in the following
experimental examples.
Experimental Examples of Ultrafine Etching
[0069] In order to be able to cover an aluminum alloy with
depressions and protrusions of such a size, it is necessary to
search for the immersion conditions by trial and error. When a
60.degree. C. aqueous solution containing a hydrazine monohydrate
concentration of 3.5% is used to immerse A5052 and A7075 stock,
setting the immersion time to about 2 minutes is optimal. The
surface obtained from this immersion time is covered entirely with
depressions having a diameter of 10 to 100 nm, the number-average
diameter being 40 to 50 nm. However, when the stock is immersed for
4 minutes, the diameter of the depressions rises to 80 to 200 nm,
the number-average diameter is increased suddenly to more than 100
nm and additional depressions are formed at the bottom of these
pits, making the structure more complex. In addition, when the
stock is immersed for 8 minutes, erosion in the form of lateral
holes also proceeds, resulting in somewhat of a sponge-like shape,
in addition to which deep depressions are interconnected, changing
into a valley or ravine-like shape. With 16 minutes of immersion,
the aluminum alloy visibly changes from the original metal color,
becoming somewhat brownish, from which it is apparent that the
manner in which visible light is absorbed has begun to change.
[0070] When the immersion time under the above conditions was one
minute, depressions having a diameter of 10 to 40 nm were observed
with electron micrographs. The number-average diameter of these
depressions was 25 to 30 nm. In addition, with 0.5 minute of
immersion, the depressions covering the surface had a diameter of
10 to 30 nm, the number-average diameter for these depressions
being about 25 nm, which is not so different from the results for
an immersion time of one minute. Electron micrographs of an
aluminum alloy immersed for 0.5 minute and an aluminum alloy
immersed for one minute were carefully compared, whereupon the
depth of the depressions was clearly shallower in the alloy
immersed for 0.5 minute than in that immersed for one minute. Thus,
in A5052 and A7075 stock immersed in a weakly basic aqueous
solution, erosion begins at a period of 20 to 25 nm, with
depressions having a diameter of 20 nm first being formed for some
unknown reason. When the depth of these depressions has been
increased to about the same level as the diameter, the edges of the
depressions are then eroded, the depression diameter is increased
and erosion in indefinite directions at the interior of the
depressions begins. When erosion has taken place in this way, the
simple and robust erosion that is most suitable for adhesive
joining was achieved with immersion of the A7075 or A5052 in a 3 to
5%-aqueous solution (60.degree. C.) of hydrazine monohydrate for
about 2 minutes.
[0071] A resultant obtained, for example, by use of the one-pack,
high-temperature curing epoxy adhesive EP106 (available from
Cemedine, Japan) having a viscosity of 40 Pas at a temperature of
23.degree. C., will be described. Based on the results of the
adhesion experiments shown in the working examples, in the case of
an aluminum alloy stock such as A7075 that was immersed for one
minute in an aqueous solution of hydrazine hydrate under the
conditions mentioned above, the fine depressions had a
number-average diameter of about 25 nm, which was too small and the
epoxy resin appeared to have difficulty in entering into these fine
depressions. The joining strength apparently reached a maximum
value in cases where the immersion time was set to two minutes.
When A7075, etc. was immersed for two minutes under the
above-mentioned conditions, the fine depressions had a
number-average diameter of about 40 nm; this epoxy resin would
presumably have been able to make its way into the interior of fine
depressions of at least about this size.
[0072] In short, when the inner surfaces of micron-order
depressions are textured faces having irregularities with
depressions and protrusions at a period of several tens of
nanometers, the joining strength rises. Also, when the
above-described immersion time is longer than 2 minutes, e.g., when
it is extended to 4 minutes or 8 minutes, not only is the
depression diameter increased but also other depressions are formed
within the depressions, resulting in what could be characterized as
a sponge-like shape. Thus, not only is the strength of the aluminum
alloy surface layer itself weakened but also the adhesive is unable
to penetrate all the way into the back of the complex holes. As a
result, the number of voids at the joining interface of the joined
object is increased, resulting in a joining strength that is lower
than the maximum value. In short, when the above-mentioned epoxy
adhesive is used for an aluminum alloy such as A7075, in addition
to providing a suitable micron-order roughness, it is desirable to
cover the surface with ultrafine depressions having a
number-average diameter of 40 to 50 nm, in order to maximize the
joining strength. It can be appreciated that the range in the
optimal immersion time for creating such ultrafine depressions is
very narrow. This is because the best joining results were obtained
at an immersion time of about 2 minutes as mentioned above.
[0073] When the same epoxy adhesive was used on A5052 aluminum
alloy, the immersion conditions during etching in an aqueous
solution of caustic soda differed somewhat from those for A7075.
Obviously, this is probably due to differences in the manner of
erosion and in the properties of the eroded surface. Further
details will be given in the subsequently described experiment
examples.
[0074] Ammonia water has a lower pH than an aqueous solution of
hydrazine and, when the aqueous solution is set to a temperature
higher than standard temperature, evaporation of the ammonia
becomes vigorous. For these reasons, such immersion treatment is
carried out at a high concentration and a low temperature; even
when very strong ammonia water having a concentration of about 25%
is used, an immersion time of 15 to 20 minutes is required.
Conversely, most water-soluble amines form aqueous solutions having
a stronger basicity than aqueous solutions of hydrazine, so the
period of treatment will be shorter. In mass production treatment,
the stability of the operation is lost if the immersion time is too
long or too short. In this sense, the use of hydrazine hydrate,
which allows the optimal immersion time to be set to several
minutes, appears to be suitable for practical purposes.
[0075] In all cases, there were types of alloys for which the
joining strength was improved upon immersion in an aqueous solution
of hydrogen peroxide of several percent concentration following
immersion in an aqueous solution of hydrazine hydrate, ammonia or a
water-soluble amine. The thickness of the metal oxide layer on the
surface may be increased but analysis is difficult at thicknesses
greater than 2 nm, making theoretical elucidation impossible.
[Epoxy Resin-Based Adhesives and their Application]
[0076] A commercially available epoxy resin for adhesives may be
used as the initial material for the epoxy-based adhesive employed
in the present invention. Commercially available epoxy resins for
adhesives include, for example, bisphenol-type epoxy resins,
glycidyl amine-type epoxy resins, polyfunctional polyphenol-type
epoxy resins and alicyclic epoxy resins. Any of these types may be
used in the present invention. Alternatively, use may be made of a
product obtained by coupling together these epoxy resins via
reaction with a polyfunctional third ingredient, such as a
polyfunctional oligomer having a plurality of hydroxyl groups. It
is preferable to add a polyfunctional amine compound as a curing
agent to these epoxy resins for adhesives and mix those together so
as to create a one-pack epoxy adhesive.
Elastomer Ingredient
[0077] Adding an elastomer ingredient or the like to the
above-mentioned epoxy-based adhesive ingredients is desirable in
that a cushioning medium can be obtained for the situation where
the joint is subjected to temperature shock or physical shock.
Also, the addition of filler ingredients to the above ingredients
is advantageous for enabling the linear expansion coefficient to be
matched to that of the aluminum alloy and that of the CFRP
material. The elastomer ingredient is preferably included in an
amount of from 0 to 30 parts by weight per 100 parts by weight of
the above resin ingredients combined (epoxy resin
ingredients+curing agent ingredients). Including an excessive
amount of the elastomer ingredient of more than 30 parts by weight
will lower the joining strength, which is undesirable. One of the
elastomer ingredients is a vulcanized rubber powder having a
particle size of from several microns to 15 .mu.m. Even when the
adhesive is applied, the particles with a particle size of several
microns or more are somewhat too large to enter the depressions on
the aluminum alloy. As a result, they remain entirely in the
adhesive layer and have no influence on the anchoring areas.
[0078] As a result, this ingredient has a role of withstanding
shock without a loss of joining strength. Various types of
vulcanized rubber may be used, although, regardless of the type of
rubber, milling to a micron-order particle size is difficult. To
the extent of the investigations conducted by the inventors,
perhaps owing to a lack of need, there appears to be no sign of any
research and development having been actively carried out on
methods for manufacturing vulcanized rubber in the form of fine
particles. The inventors tried the method of cooling a natural
rubber vulcanizate (an ingredient which is substantially the same
as the ingredient formulated in tire rubber, excluding carbon
black) with liquid nitrogen, followed by mechanical milling and
classification. Unfortunately, the product costs are not within a
realm that is commercially viable. Another approach was to use
unvulcanized or semi-crosslinked rubber and a modified super
engineering plastic or a polyolefin resin. An exemplary modified
super engineering plastic is the hydroxyl-terminated polyether
sulfone PES100P (available from Mitsui Chemical, Japan). That is,
use may be made of end-modified thermoplastic resins of a
hydroxyl-terminated polyether sulfone which have a melting point or
softening point of at least 300.degree. C.
[0079] Polyolefin resins which mix easily with epoxy resins have
already been developed and may also be advantageously used. The
inventors believe the durability against temperature shock to be
theoretically inferior to that of powdered vulcanized rubber but
this is not yet in fact well known. The evaluation techniques
themselves have not been carried out to the limit in the methods by
the inventors. In any case, even the admixture of such unvulcanized
elastomers was found to result in a high strength to temperature
shock. Examples of such polyolefin resins include maleic
anhydride-modified ethylene copolymers, glycidyl
methacrylate-modified ethylene copolymers, glycidyl ether-modified
ethylene copolymers and ethylene alkyl acrylate copolymers.
Filler
[0080] In addition, the filler will be described. It is preferable
to use an epoxy adhesive composition which additionally contains
from 0 to 50 parts by weight of filler per 100 parts by weight of
total resin ingredients including elastomer ingredient. Examples of
fillers include reinforcing fiber-type fillers such as carbon
fibers, glass fibers and aramid fibers; and powder-type fillers
such as calcium carbonate, mica, glass flakes, glass balloons,
magnesium carbonate, silica, talc, clay and crushed carbon fibers
or aramid fibers. Carbon nanotubes may also be used as filler.
Preparation of Epoxy Adhesive
[0081] Preparation of the epoxy adhesive entails thoroughly mixing
together the main epoxy resin ingredients, elastomer, filler and
curing agent and also, depending on the viscosity, mixing a solvent
for epoxy adhesives (which is also commercially available). In the
present invention, this mixture serves as an adhesive composition
(an uncured epoxy resin composition). This adhesive composition is
applied to the necessary places on the aluminum alloy part obtained
in the preceding step. Either manual application with a brush or
application with an adhesive coating equipment may be adopted.
[Treatment Step Following Application of Epoxy Resin Adhesive]
[0082] After coating, it is preferable to place the coated piece in
a vacuum vessel or a pressure vessel, reduce the pressure almost to
vacuum, hold the piece in this state for several minutes and then,
by introducing air, either return the pressure to standard pressure
or apply a pressure of several atmospheres or several tens of
atmospheres. In addition, repeated cycling with raising and
lowering of pressure is also advantageous. In this way, air and
gases between the coating material and the aluminum alloy are
removed, facilitating penetration of the applied material into the
ultrafine depressions. In actual mass production, using a pressure
vessel and high-pressure air leads to increased costs, both in
respect of equipment and operating expenditures, so it is
economically effective to carry out one or several cycles of
pressure reducing/standard pressure restoration or pressure
reducing/pressurization to several atmospheres using a vacuum
vessel. Using the aluminum alloy according to the present
invention, a sufficiently stable joining strength can be achieved
with several cycles of pressure reducing/normal pressure
restoration. It is also desirable to then remove the workpiece from
the vessel and let it to stand for several hours in an environment
of normal temperature or at approximately 40.degree. C. Even in
cases where some solvent has been added to the epoxy adhesive
composition, this will enable most of the solvent to be vaporized
off.
[FRP Prepreg]
[0083] Fiber-reinforced plastics (FRP) are exemplified by glass
fiber-reinforced plastics (abbreviated below as GFRP), aramid
fiber-reinforced plastics (AFRP) and CFRP. In this specification,
FRPs in which epoxy resin serves as the matrix are discussed, while
any type of FRP may be employed in the present invention, including
those using glass fibers, carbon fibers or aramid fibers, as well
as those which use other reinforcing fibers. The description that
follows relates specifically to CFRPs, in which the effects of the
present invention are most readily apparent. On the other hand,
specific description of other FRPs will be omitted.
[0084] Use of a commercially available CFRP prepreg may be done, of
course. Commercially available products sold as prepregs include
carbon fiber woven cloth impregnated with the above-mentioned
epoxy-based adhesive and prepregs produced by creating a film-like
material from the above-mentioned epoxy resin in uncured form and
laminating the material with a carbon fiber woven cloth. Such
techniques are known in common and any person with ordinary skill
in the art, not limited only to specialized manufacturers, could
easily fabricate a prepreg using a carbon fiber woven fabric and a
one-pack epoxy-based adhesive. Epoxy resins commonly used for this
purpose include dicyandiamide-curable epoxy resins and
amine-curable epoxy resins. Such resins are designed to maintain a
B stage (a nearly solid but uncured state) at normal temperature,
melt in the course of temperature rise to somewhere in excess of
100.degree. C. and be cured thereafter. In this sense, it is
preferable that the epoxy-based adhesive coated onto the aluminum
alloy part and the epoxy-based uncured resin (adhesive) used in the
CFRP prepreg have identical curing temperature properties. However,
in experiments by the inventors, a high joining strength was
obtained even with a prepreg that was heat-cured without particular
adjustment of these curing temperature properties. It may thus be
possible to obtain even better united object by carrying out
further detailed investigations.
[0085] The uncured prepreg is cut into the required shape and
stacked in the required manner to prepare the prepreg portion. That
is, when a plurality of sheets of a unidirectional prepreg (a
prepreg composed of a cloth woven from many warp threads but very
few filling threads) are stacked to form a plate-like laminate, the
strength directionality of the finished laminate can be controlled,
by stacking the fabric with the filaments oriented in the same
direction or at differing angles. In this, there appears to be much
know-how on the assembly of such prepregs. Also, it is reported
that, the number of warp threads and the number of filling threads
is the same in a carbon fiber plain-woven cloth and, when prepregs
are stacked while changing the angle 45 degrees in each successive
layer, the final CFRP laminate will have the same strength in all
directions. Hence, the required number of layers and the manner of
stacking are planned preliminarily, then each prepreg is cut and
stacked according to the plan to complete the laminate.
[Method for Layering the Prepregs and Manufacturing the
Composite]
[0086] The above FRP prepreg (plate-like laminate) is placed on the
earlier described epoxy adhesive-coated aluminum alloy part. When
heated in this state, the epoxy resin adhesive and the epoxy resin
in the prepreg melt, following which they are cured. For strong
joining, it is necessary for heating to be carried out in a state
where the prepreg and the aluminum alloy part are pressed together
and for air present therebetween to be expelled during melting of
the resin. For example, by prefabricating a seat which matches the
shape of the face of the aluminum alloy on the side opposite to the
joining face thereof, laying some aluminum foil or polyethylene
film down on the seat, then setting the aluminum alloy part on the
foil or film, placing the prepreg on the aluminum alloy part,
laying polyethylene film on top of the prepreg, placing thereon a
separately fabricated member to be fixed thereto that was
separately fabricated from structural stock or the like and matched
to the prepreg shape of the finished product and setting a weight
on top, pressing and fixing can be carried out during hot
curing.
[0087] Of course, any of various methods other than gravity alone
may be utilized to induce curing while pressing both sides
together. In the case of a member of aircraft, the entire assembly
described above is enclosed in a film bag, then overheated while
lowering the pressure, thereby forcibly drawing out air from the
interior when all the epoxy component has melted. Because the
prepreg tightens when the air has been drawn out to a certain
degree, air is subsequently supplied into the film bag, thereby
inducing curing under increased pressure in this arrangement.
Experiments were carried out here under the assumption that the air
within the prepreg is drawn off substantially by the mechanical
pressure applied during melting of the epoxy component.
[0088] Heating is carried out by placing the above overall assembly
in a hot-air dryer or an autoclave. Generally it is preferable to
hold the assembly at between 110 and 140.degree. C. for several
tens of minutes, thereby causing the adhesive component to melt and
inducing gelation, then raise the temperature to between 150 and
170.degree. C. and heat for another several tens of minutes to
effect curing. The optimal temperature conditions vary depending on
the epoxy component and the curing agent component. After curing,
the assembly is allowed to be cooled, then the metallic mold is
removed and the shaped object is taken out. When aluminum foil or
polyethylene film are used as described above, these are peeled off
to enable the object to be released from the mold.
Method of Measuring Joining Strength
[0089] Measurement of the joining strength between the aluminum
alloy and the FRP used in the present invention will be described
below. FIG. 1 is a sectional view of a firing jig for joining the
aluminum alloy and the FRP together. FIG. 2 shows a test piece of
an aluminum alloy composite 10 fabricated by firing the aluminum
plate and CFRP in the firing jig 1. The firing jig 1 is a holder
for immobilizing the aluminum alloy plate 11 and the prepreg 12
when they are fired. The metallic mold body 2 is open at a top
surface thereof and has formed therein a recessed portion 3 of
rectangular shape. A mold throughhole 4 is formed at the bottom
thereof.
[0090] A bottom plate projection 6 on a mold bottom plate 5 is
inserted into the mold throughhole 4. The bottom plate projection 6
extends out from the mold bottom plate 5 on the mold body 2. The
mold body 2 is mounted on top of a mold seat 8 with the bottom face
7 mating with the seat 8. In the situation where the mold bottom
plate 5 is inserted and placed within the recessed portion 3 of the
mold body 2, an aluminum alloy composite 10 composed of an aluminum
alloy plate 11 joined with CFRP 12 as shown in FIG. 2 is produced
through firing. Briefly, production of this aluminum alloy
composite 10 is carried out by the following procedure. First, a
mold release film 17 is laid down over the entire upper face of the
mold bottom plate 5. An aluminum alloy plate 11 and a plate-like
PTEF spacer 16 are set on top of the mold release film 17. A
required prepreg 12 is placed on this PTEF spacer 16 and along with
it on top of an end of the aluminum alloy plate 11. The prepreg 12
is a CFRP prepreg composed of a reinforcing fiber woven cloth
impregnated with an uncured epoxy-based adhesive.
[0091] After laying down the prepreg 12, a piece of mold release
polyethylene film 13 is additionally placed on both the aluminum
alloy plate 11 and the prepreg 12. Then, PTEF blocks 14 and 15 of
PTEF (polytetrafluoroethylene resin) are set on top thereof as
weights. In addition, where necessary, weights of several hundred
grams (not shown) are placed on top of the blocks. The resulting
assembly is loaded in this state into a firing oven, where the
prepreg is cured, then allowed to be cooled. The weights, seat 8
and the like are then removed and, by pushing the bottom end of the
bottom plate projection 6 against the surface of the floor below,
the aluminum alloy composite 10 (see FIG. 2) obtained by joining
the aluminum plate with the CFRP can be removed together with the
mold release films 13 and 17. The PTEF spacer 16 and the mold
release films 17 and 13, being made of non-stick materials, can
then be easily peeled from the CFRP.
[Examples of Methods for Using Composites]
[0092] FIG. 3 is a partial solid view showing an example in which
an aluminum alloy/FRP composite according to the present invention
and a metal structural material (angle bar) are joined together
using bolts and nuts. The aluminum alloy composite 20 is a
composite obtained by uniting an aluminum alloy with CFRP. The CFRP
21 is a plate-like structure produced by firing the prepreg. The
angle bar 23 is a prefabricated structural material. Rectangular
reinforcing plates 22 are integrally joined with the front and back
sides of the CFRP 21. The reinforcing plates 22 are made of
aluminum alloy (e.g., A7075) and have been fired and integrally
joined beforehand with the CFRP 21 by the above-described
method.
[0093] The CFRP 21, the reinforcing plates 22 on the front and back
sides thereof and the angle bar 23 are fastened together so that no
movement occurs therebetween by means of a bolt 25 along with a nut
placed on the underside of the angle bar 23 (not shown) through an
upper washer 24 placed on the reinforcing plate 22 and a lower
washer (not shown) placed on the bottom face of the angle bar 23.
In the aluminum alloy composite 20 obtained by joining together
rectangular plates 22 made of aluminum alloy (A7075) and CFRP 21,
the joining strength between both materials is extremely high, the
shear breaking strength being from 50 to 70 MPa. Moreover, the
forces applied to the plates 22 by the bolt 25 and washer 24 or the
like can be effectively dispersed on the CFRP 21. In short, even
when the bolt 25 and nut are tightened with sufficient strength,
only the plates 22 made of A7075 are deformed; the CFRP 21 in the
composite 20 is not damaged. An aluminum alloy and CFRP can be
strongly joined by the method for manufacturing an aluminum alloy
composite according to the present invention as stated above.
[0094] As described in detail above, in the aluminum alloy
composite and method for manufacturing thereof according to the
present invention, an aluminum alloy part and FRP are strongly
united, thereby enabling tough, lightweight components to be
produced. Because parts obtained by integrally uniting an aluminum
alloy member with an FRP member can be manufactured as parts for
aircrafts and other applications, novel constructions may be
employed when the need arises to join and fasten both
materials.
[0095] Also, because the aluminum alloy composite according to the
present invention is light, it can be employed in components for
vehicles, such as parts for automotives and parts for bicycles,
where lighter weight is strongly desired. That is, because the
aluminum alloy stock can be formed in a relatively free shape by
mechanical working or the like, existing fastening methods such as
a bolt-and-nut technique or a screwing technique can be employed to
join two pieces of aluminum alloy. FRP parts, on the other hand,
are suitable for forming objects with plate-like or pipe-like
shapes rather than complex shapes and even make it possible to
create large or elongated cured objects. As a result, composite
structures can be produced which exhibit the characteristics of
both materials.
[0096] Hence, by having the end of a integrally united part made of
aluminum alloy, it is possible to arrive at a member capable of
being easily assembled with nuts and bolts or with screws. In
summary, by precisely designing the surface of the aluminum alloy,
the joining strength and accuracy in joining with the epoxy resin
can be made vastly high, making it possible to employ new methods
of fabrication and assembly that utilize such joining strength.
[0097] Embodiments of the present invention will be described below
by means of experiment examples. FIG. 4 shows the configuration in
which two pieces of aluminum alloy are joined together with an
adhesive (metal on metal). In the figure, aluminum alloy piece 30
and aluminum alloy piece 31 are aluminum alloy plates made of the
same material respectively. Aluminum alloy piece 30 and aluminum
alloy piece 31 are joined at a joining face 32. At this joining
face 32, as described above, a micron-order surface roughness is
formed by chemical etching at the surface. This surface has a thin
layer of sodium ion-free aluminum oxide with a thickness of at
least 2 nm and the interior of the surface roughness is covered by
an ultra fine irregular surface with depressions and protrusions
having a diameter of 10 to 100 nm and the same depth or height.
[0098] It is to these ultrafine irregular surfaces that joining has
taken place through an intervening epoxy-based adhesive. FIG. 2
shows a test piece for measuring the joining strength obtained when
an aluminum alloy plate and FRP are joined together with an
epoxy-based adhesive. As explained above, FIG. 3 is a schematic
view for a case in which a composite obtained by joining an
aluminum alloy plate and FRP with epoxy resin is connected with
another metal part by means of bolts and nuts. Specific working
examples will be described subsequently. The following instruments
were used for obtaining measurements.
(a) X-Ray Surface Examination (X-Ray Photoelectron Spectroscopy
(XPS))
[0099] An ESCA spectrometer (AXIS-Nova, manufactured by
Kratos/Shimadzu Corporation, Japan) was used to examine the
constituent elements on a surface of several micrometers in
diameter to a depth of 1 to 2 nm.
(b) Electron Microscopic Examination
[0100] Observation was carried out at 1 to 2 kV using scanning
electron microscopes (SEM) (S-4800, manufactured by Hitachi, Ltd.,
Japan) and JSM-6700F (JEOL, Japan).
(c) Scanning Probe Microscopic Examination
[0101] A SPM-9600 (Shimadzu Corporation, Japan) was used.
(d) X-Ray Diffraction Analysis (XRD Analysis)
[0102] An XRD-6100 (Shimadzu Corporation, Japan) was used.
(e) Measurement of Joining Strength in Composite
[0103] Using a tensile testing machine (Model 1323, manufactured by
Aikoh Engineering, Japan), the shear breaking strength was measured
at a tension speed of 10 mm/min.
WORKING EXAMPLES
Experiment Example 1
Adhesion with A5052 Aluminum Alloy
[0104] Commercially available A5052 sheet stock having a thickness
of 1.6 mm was acquired and cut to form a plurality of rectangular
pieces measuring 45.times.18 mm. Water was made ready in a tank and
a commercially available aluminum alloy degreaser (NE-6, available
from Meltex, Japan) was added to the water to form a aqueous
solution of 60.degree. C. having a concentration of 7.5%. The
rectangular pieces of aluminum alloy were immersed in the solution
for 7 minutes and then thoroughly rinsed with water. An aqueous
solution of hydrochloric acid having a concentration of 1% and
adjusted to 40.degree. C. was made ready in another tank, in which
the aluminum alloy pieces were immersed for 1 minute and then
thoroughly rinsed with water. Next, an aqueous solution of caustic
soda having a concentration of 1.5% and adjusted to 40.degree. C.
was made ready in still another tank, in which the same aluminum
alloy pieces were immersed for 2 minutes and then thoroughly rinsed
with water. An aqueous solution of nitric acid having a
concentration of 3% and adjusted to 40.degree. C. was made ready in
yet another tank, in which the aluminum alloy pieces were
subsequently immersed for 1 minute and then rinsed with water.
[0105] Next, an aqueous solution containing hydrazine monohydrate
by 35% and adjusted to 60.degree. C. was made ready in another
tank, in which the aluminum alloy pieces were immersed for 1
minute, rinsed with water and then dried by being placed in a
hot-air dryer set at 67.degree. C. for 15 minutes. After drying,
the aluminum alloy pieces were collected together, wrapped with
aluminum foil and then placed in a plastic bag, which was closed
and stored. Four days later, one of the pieces was examined with an
electron microscope and found to be covered with depressions of
diameters ranging from 10 to 80 nm and a number-average diameter of
25 to 30 nm.
[0106] This is shown in the micrograph of FIG. 9. Another piece was
examined with a scanning probe microscope, with which the piece was
scanned over a length of 20 .mu.m at 40 .mu.m/s, and the mean peak
spacing (RSm) and maximum height (Rz) according to JIS B 0601: 2001
were measured, yielding respective results of 1.1 .mu.m (RSm) and
0.3 .mu.m (Rz). FIG. 15 shows these roughness curves. In addition,
still another piece was subjected to examination with XPS and
aluminum atoms were observed. In contrast with the purchased A5052,
which was observed to have a ratio of zero-valent aluminum to
trivalent aluminum of about 1:3, the piece, for which surface
treatment was carried out, showed no sign of zero-valent aluminum
and the thickness of the aluminum oxide layer had increased.
Because the atomic composition to a depth of up to 1 to 2 nm from
the surface was detected in XPS, it was apparent that the aluminum
oxide surface layer had a thickness of more than 2 nm.
[0107] The same day, ten pieces of aluminum alloy for which surface
treatment had been performed as described above were taken out and
lightly coated on the ends with a commercially available
liquid-type one-pack dicyandiamide-curable epoxy adhesive (EP-106,
available from Cemedine, Japan). The pieces were placed in a
desiccator with the coated side facing upward, the desiccator was
evacuated with a vacuum pump to 1 mmHg and left in this state for
one minute, following which air was introduced, returning the
interior to standard pressure. This operation of reducing the
pressure and then returning to standard pressure was repeated three
times, following which the pieces were removed from the desiccator.
The pieces were transferred to the interior of a hot-air dryer and
assembled into five pairs by stacking two pieces with the
adhesive-coated sides mating with one another so that each joined
face area was about 0.5 cm.sup.2, a weight of 500 g was placed at
each joined face, the door was closed and the temperature was
rapidly raised to 135.degree. C. Forty minutes later, the hot-air
dryer setting was changed to 165.degree. C. and the rise in
temperature was awaited. Once the temperature reached 165.degree.
C., this state was held for 20 minutes, then the power on the
hot-air dryer was turned off, the door of the dryer was opened and
the interior was allowed to be cooled. By means of this operation,
as shown in FIG. 4, united test pieces composed of a first aluminum
alloy piece 30 and a second aluminum alloy piece 31 joined at the
joining face 32 respectively were obtained. Two days later, the
test pieces were subjected to tensile breaking tests, whereupon the
average shear strength at break for the five pairs was 48 MPa,
which was very strong.
Experiment Example 2
Adhesion with A5052 Aluminum Alloy
[0108] Surface treatment for pieces of A5052 plate was carried out
in the same way as Experiment example 1. Up through treatment with
the degreaser (NE-6, available from Meltex, Japan), preliminary
pickling treatment with a 1%-aqueous solution of hydrochloric acid
and alkali etching treatment with a 1.5%-aqueous solution of
caustic soda, treatment was carried out in exactly the same as in
Experiment example 1. Surface treatment ended at this point and the
pieces were stored in the resulting state. The same day, these
aluminum alloy pieces were taken out, the epoxy adhesive EP-106 was
lightly coated on the ends and desiccation treatment, joining and
curing of the adhesive in a hot-air dryer were subsequently carried
out in exactly the same way as in Experiment example 1, yielding
the test pieces shown in FIG. 4.
[0109] In tensile failure tests carried out two days later, the
average shear strength at break for the four pairs was 48 MPa.
These test pieces thus exhibited the same high joining strength as
in Experiment example 1. The aluminum alloy pieces following
surface treatment were subjected to XPS analysis, where large peaks
for oxygen and aluminum were observed and small peaks for
magnesium, zinc and sodium were also noted. In this experiment
example, in which surface treatment ended with treatment in an
aqueous solution of caustic soda, it was shown that sodium ions are
not removed by rinsing with water alone.
Experiment Example 3
Adhesion with A5052 Aluminum Alloy
[0110] Pieces of A5052 plate were treated in the same way as
Experiment example 1 up to a given point. Up through treatment with
the degreaser NE-6, preliminary pickling treatment with a
1%-aqueous solution of hydrochloric acid, alkali etching treatment
with a 1.5%-aqueous solution of caustic soda and neutralization
treatment with a 3%-aqueous solution of nitric acid, treatment was
carried out in exactly the same as in Experiment example 1. Surface
treatment ended at this point and the pieces were stored in this
state.
[0111] When these aluminum alloy pieces were subjected to XPS
analysis, peaks for sodium were not observed, even cumulatively.
Hence, sodium was concluded to have been removed by immersion in
the aqueous nitric acid solution and rinsing with water. The same
day, these aluminum alloy pieces were taken out, the epoxy adhesive
EP-106 was lightly coated on the ends and desiccation treatment,
joining and curing of the adhesive in a hot-air dryer were
subsequently carried out in exactly the same way as in Experiment
example 1. In tensile breaking tests carried out two days later,
the average shear strength at break for the four pairs was 44 MPa.
The joining strength was somewhat lower than in Experiments 1 and
2.
Experiments Example 4 to 12
Adhesion with A5052 Aluminum Alloy
[0112] In Experiment examples 4, 5 and 6, surface treatment up to
immersion in an aqueous solution of hydrazine hydrate was carried
out in the same way as in Experiment 1. Compared with Experiment
example 1, only the immersion times in the aqueous solution of
caustic soda were made different. These immersion times were 1
minute in Experiment example 4, 2 minutes in Experiment example 1,
4 minutes in Experiment example 5 and 8 minutes in Experiment
example 6. In Experiment examples 7, 8 and 9, surface treatment up
to immersion in an aqueous solution of caustic soda was carried out
in the same way as in Experiment example 2. Compared with
Experiment 2, only the immersion times in the aqueous solution of
caustic soda were made different. These immersion times were 1
minute in Experiment example 7, 2 minutes in Experiment example 2,
4 minutes in Experiment example 8 and 8 minutes in Experiment
example 9.
[0113] In Experiments examples 10, 11 and 12, surface treatment up
to immersion in an aqueous solution of nitric acid soda was carried
out in the same way as in Experiment example 3. Compared with
Experiment example 3, only the immersion times in the aqueous
solution of caustic soda were made different. These immersion times
were 1 minute in Experiment example 10, 2 minutes in Experiment
example 3, 4 minutes in Experiment example 11 and 8 minutes in
Experiment example 12. The results for Experiment examples 1 to 12
are shown in Table 1. The same results are plotted as graphs in
FIG. 13. The ordinate represents the adhesive strength between the
A5052 aluminum alloy pieces and the abscissa represents the
immersion time in a 1.5%-aqueous solution of caustic soda.
TABLE-US-00001 TABLE 1 Shear strength at break of A5052. Immersion
time in 1.5%-caustic soda solution Treatment step 1 minute 2
minutes 4 minutes 8 minutes (1) Up to Experiment Experiment
Experiment Experiment caustic soda example 7 example 2 example 8
example 9 treatment Joining 32 Mpa 48 Mpa 45 Mpa 35 Mpa strength
(2) Up to Experiment Experiment Experiment Experiment
neutralization example 10 example 3 example 11 example 12 treatment
Joining 20 Mpa 44 Mpa 41 Mpa 40 Mpa strength (3) Up to Experiment
Experiment Experiment Experiment hydrazine example 4 example 1
example 5 example 6 treatment Joining 29 Mpa 48 Mpa 45 Mpa 36 Mpa
strength
[0114] As is apparent from FIG. 13 and Table 1, using a combination
of A5052 treated with a 1.5%-aqueous solution of caustic soda and
epoxy resin composition EP-106, the highest joining strength is
obtained at an immersion time of about 2 minutes in the aqueous
solution of caustic soda. What is distinctive here is that, among
the test pieces which were immersed for 2 minutes in an aqueous
solution of caustic soda, the product etched with caustic soda
(Experiment example 2) and the product obtained by subsequent
neutralization as well as hydrazine treatment (Experiment example
1) exhibited substantially the same level of joining strength. Of
the treated products in which sodium ions were not present, those
subjected to treatment up to hydrazine treatment exhibited the
highest adhesive strength.
[0115] Also, in cases where the immersion time in an aqueous
solution of caustic soda was 8 minutes, etc., when the A5052
aluminum alloy pieces were etched too long, the product treated up
to neutralization (Experiment example 12) had the highest joining
strength. In comparing the product obtained by surface treatment up
to immersion in a caustic soda solution with the product obtained
by surface treatment up to immersion in a hydrazine solution, the
results were reversed. Long immersion in an aqueous solution of
caustic soda resulted in a roughness periodicity of several
microns, with little difference in the period itself. However, the
depth and height difference were changed drastically (based on
examination with scanning probe microscope) and lateral holes even
formed, resulting in an excessively complex surface shape which was
thought to have weakened the joining strength.
[0116] Because neutralization treatment is a step that actually
involves immersion in an acidic aqueous solution, aluminum alloy,
being an amphoteric metal, dissolves slightly even in acidic water;
in particular, protruding portions reportedly dissolve easily
particularly in an acidic aqueous solution. Therefore, it is
considered that Experiment example 12 exhibited a better adhesive
strength than in Experiment example 9 because the excessively
complex roughness was leveled out somewhat, returning to a suitable
degree of roughness.
Experiment Examples 13 to 15
Adhesion with A5052 Aluminum Alloy
[0117] Joining/breaking experiments were carried out in the same
way as in Experiment example 1. However, unlike in Experiment
example 1, the immersion time in the 1.5%-aqueous solution of
caustic soda was set to 2 minutes, the best time in Experiment
examples 1 to 12 and the immersion time in the 3.5%-aqueous
solution of hydrazine hydrate was varied in the range of 1 to 8
minutes.
[0118] That is, the immersion time was set to 1 minute in
Experiment example 1, 2 minutes in Experiment example 13, 4 minutes
in Experiment example 14 and 8 minutes in Experiment example 15.
The results are shown in Table 2 and FIG. 14.
TABLE-US-00002 TABLE 2 Shear strength at break of A5052 (with 2
minutes immersion in 1.5%-aqueous caustic soda) Hydrazine hydrate
immersion time 1 minute 2 minutes 4 minutes 8 minutes Experiment
Experiment Experiment Experiment example 1 example 13 example 14
example 15 Joining 48 Mpa 60 Mpa 41 Mpa 28 Mpa strength
[0119] As shown in Table 2 and FIG. 14, the highest joining
strengths were obtained with an immersion time of 2 minutes in an
aqueous solution of hydrazine hydrate. Examination of the surface
with an electron microscope showed that the entire surface was
covered with depressions having a diameter ranging from 20 to 100
nm and a number-average diameter of 40 to 50 nm. A joining force of
60 MPa (about 600 kgf/cm.sup.2) corresponds to a breaking force of
300 kgf for an object having a joining face area of 0.5 cm.sup.2.
FIGS. 9, 10, 11 and 12 show electron micrographs of the A5052
aluminum alloys obtained in Experiment examples 1, 13, 14 and 15 at
magnifications of 10,000 times and 100,000 times respectively. It
can be seen from these that, as the immersion time increases, the
finely textured surface becomes more complex, in this case the
diameter of the fine depressions becomes larger and other
depressions are formed within the earlier formed depressions,
leading to a rapid decline in the strength of the surface layer
itself on the aluminum alloy.
[0120] It will be appreciated that, on a surface like that shown in
the electron micrograph of FIG. 12, a viscous epoxy-based adhesive
will be unable to penetrate deep into the complex fine depressions.
Further, on examining the same aluminum alloy piece as used in
Experiment example 13, for which the highest joining strength was
exhibited, with a scanning probe microscope, the mean roughness
period (this is the same as the mean peak spacing RSm according to
JIS) was 1.2 .mu.m and the maximum height (Rz) was 0.5 .mu.m.
Experiment Example 16
Adhesion with A5052 Aluminum Alloy
[0121] As in Experiment example 13, pieces of A5052 sheet stock
were degreased and rinsed with water, subjected to preliminary
pickling and rinsed with water, etched with an aqueous solution of
caustic soda and rinsed with water, neutralized with an aqueous
solution of nitric acid and rinsed with water, then finely etched
with an aqueous solution of hydrazine monohydrate and rinsed with
water. Next, the pieces for which surface treatment was performed
were placed in a hot air dryer set to 150.degree. C. for 15 minutes
and dried. That is, while the drying temperature in Experiment
example 13 was set to 67.degree. C., the wet A5052 following
rinsing was directly loaded at an elevated temperature of
150.degree. C. and dried in this experiment example. This is the
difference with Experiment example 13.
[0122] Exactly the same procedure as in Experiment example 1 was
subsequently followed. Using the epoxy adhesive EP-106, two pieces
of aluminum alloy at a time were adhesively joined with each other.
On the following day after joining, the resulting test pieces were
subjected to tensile breaking tests, whereupon the average shear
strength at break for the four pairs was 43 MPa. This joining
strength was lower than in Experiment example 13. Guessing on the
basis of prior art, it would appear that, when a piece containing
some amine-based compound stands in a wet state at an elevated
temperature of more than 100.degree. C., boehmite formation occurs
at the surface. The inventors, suspecting that boehmite containing
hydroxyl groups is not always a substance having a tough hardness,
thought the results to be reasonable.
Experiment Examination 17
Adhesion with A5052 Aluminum Alloy
[0123] As in Experiment example 13, pieces of A5052 sheet stock
were degreased and rinsed with water, subjected to preliminary
pickling and rinsed with water, etched with an aqueous solution of
caustic soda and rinsed with water, neutralized with an aqueous
solution of nitric acid and rinsed with water, then finely etched
with an aqueous solution of hydrazine monohydrate and rinsed with
water. In addition, an aqueous solution of hydrogen peroxide having
a concentration of 5% was adjusted to 25.degree. C. and the pieces
were immersed therein for 5 minutes, then rinsed with water. The
aluminum alloy pieces were then dried by being placed in a hot-air
dryer set to 67.degree. C. for 15 minutes. Next, as in Experiment
example 1, two of the aluminum alloy pieces were adhesively joined
with each other at a time using the amine-curable epoxy adhesive
EP-106 (available from Cemedine, Japan). On the following day after
joining, the resulting test pieces were subjected to tensile
breaking tests, whereupon the average shear strength at break for
the four pairs was 62 MPa, which was very high.
Experiment Example 18
A7075 Aluminum Alloy
[0124] Commercially available A7075 sheet stock having a thickness
of 3 mm was acquired and cut to form a plurality of rectangular
pieces measuring 50.times.12 mm. Water was made ready in a tank and
the commercially available aluminum alloy degreaser NE-6 was added
to the water to form an aqueous solution of 60.degree. C. having a
concentration of 7.5%. The pieces of aluminum alloy were immersed
in the solution for 5 minutes, then thoroughly rinsed with water.
An aqueous solution of hydrochloric acid having a concentration of
1% and adjusted to 40.degree. C. was made ready in another tank and
the aluminum alloy pieces were immersed therein for 1 minute, then
thoroughly rinsed with water. Next, an aqueous solution of caustic
soda having a concentration of 1.5% was made ready in still another
tank and the same aluminum alloy pieces were immersed therein for 4
minutes, then thoroughly rinsed with water. The aluminum alloy
pieces were then dried by being placed in a hot-air dryer set at
67.degree. C. for 15 minutes. After drying, the aluminum alloy
pieces were collected together and wrapped with aluminum foil, then
placed in a plastic bag, which was closed and stored.
[0125] Four days later, one of the pieces was examined with an
electron microscope. The micrograph obtained is shown in FIG. 5. A
strange, dried seaweed-like surface having diameter of 100 to 150
nm is apparent. This was subjected to XPS analysis, whereupon large
peaks for oxygen and aluminum and small peaks for magnesium,
copper, zinc and sodium were noted. In this case, surface treatment
was completed with treatment using an aqueous solution of caustic
soda and it was apparent that rinsing with water was not sufficient
to completely remove the sodium ions.
Experiment Example 19
A7075 Aluminum Alloy
[0126] Commercially available A7075 sheet stock having a thickness
of 3 mm was acquired and cut to form a plurality of rectangular
pieces measuring 50.times.12 mm. Water was made ready in a tank and
the commercially available aluminum alloy degreaser NE-6 was added
to the water to form an aqueous solution of 60.degree. C. having a
concentration of 7.5%. The pieces of aluminum alloy were immersed
in the solution for 5 minutes, then thoroughly rinsed with water.
An aqueous solution of hydrochloric acid having a concentration of
1% and adjusted to 40.degree. C. was made ready in another tank and
the aluminum alloy pieces were immersed therein for 1 minute, then
thoroughly rinsed with water. Next, an aqueous solution of caustic
soda having a concentration of 1.5% was made ready in still another
tank and the same aluminum alloy pieces were immersed therein for 4
minutes, then thoroughly rinsed with water. Next, an aqueous
solution of nitric acid having a concentration of 3% and adjusted
to 40.degree. C. was made ready in another tank and the aluminum
alloy pieces were immersed therein for one minute, then rinsed with
water. Compared with Experiment example 18, this was an experiment
example in which the step of immersion in an aqueous solution of
nitric acid was added to the steps in Experiment example 18.
[0127] Next, the aluminum alloy pieces were dried by being placed
in a hot air dryer set at 67.degree. C. for 15 minutes. After
drying, the aluminum alloy pieces were collected together, wrapped
with aluminum foil and then placed in a plastic bag, which was
closed and stored. Four days later, one of the pieces was examined
with an electron microscope. The micrograph obtained is shown in
FIG. 6. Compared with FIG. 5, the difference was dramatic. While
the periodicity of the depressions and protrusions remained the
same, the protuberant features in the dried seaweed-like pattern
had all disappeared, resulting in a shape in which only low
protrusions remained. The surface had clearly become smoother, even
when seen at a microscopic level. Such a surface showed little
promise of a high joining strength with adhesive.
Experiment Example 20
A7075 Aluminum Alloy
[0128] Commercially available A7075 sheet stock having a thickness
of 3 mm was acquired and cut to form a plurality of rectangular
pieces measuring 50.times.12 mm. Water was made ready in a tank and
the commercially available aluminum alloy degreaser NE-6 was added
to the water to form an aqueous solution of 60.degree. C. having a
concentration of 7.5%. The rectangular pieces of aluminum alloy
were immersed in the solution for 5 minutes, then thoroughly rinsed
with water. Next, an aqueous solution of hydrochloric acid having a
concentration of 1% and adjusted to 40.degree. C. was made ready in
another tank and the aluminum alloy pieces were immersed therein
for 1 minute, then thoroughly rinsed with water.
[0129] Next, an aqueous solution of caustic soda having a
concentration of 1.5% was made ready in still another tank and the
aluminum alloy pieces were immersed therein for 4 minutes, then
thoroughly rinsed with water. An aqueous solution of nitric acid
having a concentration of 3% and adjusted to 40.degree. C. was made
ready in another tank and the aluminum alloy pieces were immersed
therein for one minute, then rinsed with water. Next, an aqueous
solution containing hydrazine monohydrate by 3.5% and adjusted to
60.degree. C. was made ready in yet another tank and the aluminum
alloy pieces were immersed therein for 2 minutes, following which
they were thoroughly rinsed with water, then dried by being placed
for 15 minutes in a hot air dryer set to 67.degree. C. Compared
with Experiment example 19, this was an experiment in which the
step of immersion in an aqueous solution of hydrazine hydrate was
added to the steps in Experiment example 19. After drying, the
aluminum alloy pieces were collected together, wrapped with
aluminum foil and then placed in a plastic bag, which was closed
and stored.
[0130] Four days later, one of the pieces was examined with an
electron microscope. The micrograph obtained is shown in FIG. 7. As
is apparent when compared with FIG. 6, a strange pattern that
resembles jumbled piles of stones or boulders of 20 to 110 nm in
diameter can be seen. Judging from FIG. 7 (with magnification of
10,000 times), the surface appears to be covered by depressions
having diameters of between 40 and 100 nm, with the number-average
diameter being 60 nm. In an XPS examination of another of the
pieces, peaks for zero-valent aluminum such as those in Experiment
example 1 were not observed.
[0131] The same day, the aluminum alloy pieces were taken out and
lightly coated on the respective ends with the commercially
available epoxy adhesive EP-106. The pieces were placed in a
desiccator with the coated side facing upward, then the desiccator
was evacuated with a vacuum pump to 1 mmHg and left in this state
for one minute, following which air was introduced, returning the
interior to standard pressure. This operation of reducing the
pressure and then returning to standard pressure was repeated three
times, following which the pieces were removed from the desiccator.
These aluminum alloy pieces were transferred to the interior of a
hot air dryer and assembled into pairs by stacking two pieces with
the adhesive-coated sides placed against one another at a time so
that each joined face had an area of about 0.5 cm.sup.2, a 500 g
weight was placed at each joined face, the door was closed and the
temperature was rapidly raised to 135.degree. C.
[0132] Forty minutes later, the hot air dryer setting was changed
to 165.degree. C. and the rise in temperature was awaited. Once the
temperature had reached 165.degree. C., this state was held for 20
minutes, then the power on the hot air dryer was turned off, the
door of the dryer was opened and the interior was allowed to be
cooled. In tensile breaking tests conducted two days later on the
resulting test pieces, the average shear strength at break for the
four pairs was 65 MPa, which was very high.
Experiment Example 21
Adhesion with A7075 Aluminum Alloy
[0133] In a similar way as in Experiment example 20, pieces of
A7075 sheet stock were degreased and rinsed with water, subjected
to preliminary pickling and rinsed with water, etched with an
aqueous solution of caustic soda and rinsed with water, neutralized
with an aqueous solution of nitric acid, cleaned with water and
supersonic vibration, again immersed in an aqueous solution of
nitric acid and rinsed with water, then finely etched with an
aqueous solution of hydrazine monohydrate and rinsed with water. In
addition, an aqueous solution of hydrogen peroxide having a
concentration of 5% was adjusted to 25.degree. C. and the aluminum
alloy pieces were immersed therein for 5 minutes, then rinsed with
water. The pieces were subsequently dried by being placed for 15
minutes in a hot air dryer set to 67.degree. C.
[0134] Four days later, one of the pieces was examined with an
electron microscope. The results are shown in FIG. 8. The
transition from FIGS. 7 to 8 is due to the aqueous hydrogen
peroxide, although how to describe the change in the images is not
an easy matter. While there is not a substantial difference in the
periodicity of the fine depressions and protrusions, FIG. 8 appears
to show a well-defined shapes without underlying features. One gets
the sense that crystallization of the aluminum oxide has proceeded.
Yet, when XRD analysis was carried out, peaks for aluminum oxide
crystal were not observed. Observation with XRD is impossible
without a certain amount of crystals; it is impossible to determine
whether the aluminum oxide was microcrystalline or was still
amorphous. Another piece was subjected to examination with a
scanning probe microscope and data on the roughness obtained.
According to the results, the mean peak spacing (RSm) was 3.5 .mu.m
and the maximum height (Rz) was 1.8 .mu.m.
[0135] Next, in a similar way as in Experiment example 20, the
aluminum alloy pieces were adhesively joined together, two pieces
at a time, using the epoxy adhesive EP-106. On the second day after
joining, tensile breaking tests were carried out, whereupon the
four pairs had an average shear strength at break of 70 Mpa, which
was very high.
Experiment Example 22
Adhesion with A7075 Aluminum Alloy
[0136] A plurality of square rod-shaped pieces measuring
3.times.4.times.18 mm were formed by cutting and mechanically
working commercially available A7075 sheet stock having a thickness
of 3 mm. Each of these pieces, with a hole of 1.5 mm in diameter
formed at an end thereof, was shaped so as to facilitate liquid
treatment, after which the pieces were used in the following
experiment. Specifically, liquid treatment was carried out in
exactly the same way as in Experiment example 21. Next, an epoxy
adhesive was applied on the end face on the side opposite to that
where the hole was formed, the pieces were placed in a desiccator
and pressure reduction/standard pressure recovery operations were
carried out in the same way as in Experiment example 21. Next, two
pieces were butted end-to-end, creating three pairs of joined
objects having lengths of about 36 mm. Specifically, with the two
pieces mutually abutting in each pair, a tacky tape was wrapped
around the vicinity of the junction to keep the two pieces from
separating and to keep the adhesive from spilling out when the
viscosity is lowered under elevated temperature. The three pairs
wrapped about with tacky tapes were placed horizontally on a plate
inside a hot air dryer and weights were set on both ends to keep
them from extending, following which heating was carried out in
exactly the same way as in Experiment example 21.
[0137] This experiment was carried out in order to measure the
tensile strength at break for pieces of A7075 stock which had been
joined with an epoxy adhesive. Tensile break tests were carried out
on the resulting joined object with the tensile testing machine,
yielding an average value of 85 kgf. Given the joining surface area
of 0.12 cm.sup.2, a tensile strength at break of 69 MPa was
obtained. The average thickness of the adhesive layer at this time
was 0.25 mm. While there may be no special significance, it was
striking to note that the shear strength at break and the tensile
strength at break were substantially identical values, both being
close to 70 MPa.
Experiment Example 23
Adhesive
[0138] A commercially available liquid single-pack
dicyandiamide-curable epoxy adhesive EP-106 (available from
Cemedine, Japan) was acquired. Separately, the ethylene-acrylic
acid ester-maleic anhydride terpolymer Bondine TX8030 (available
from Arkema, France) was acquired, freeze-fractured at a liquid
nitrogen temperature to obtain a powder that passes through a 30
.mu.m screen. In addition, glass fibers having an average fiber
diameter of 9 .mu.m and a fiber length of 3 mm (RES03-TP91,
available from Nippon Sheet Glass, Japan) were acquired and lightly
ground in a mortar. 100 g of the epoxy adhesive EP-106, 5 g of the
above powdered polyolefin resin and 10 g of the above glass fibers
were placed in a polyethylene beaker, thoroughly stirred and left
to stand for one hour, after which the contents were again stirred
to blend together the ingredients. This mixture was used as the
epoxy adhesive composition.
[0139] Aside from using the resulting adhesive composition in place
of EP-106, this experiment was carried out in exactly the same way
as Experiment example 21. Two days after curing the adhesive, the
test pieces were subjected to tensile breaking tests, whereupon the
average shear strength at break for the four pairs was 69 MPa.
Based on the results from Experiment examples 1 to 21, it is
apparent that the joining strength is basically determined by the
shape and physical properties of the metal surface. As the results
in this experiment are substantially the same as those in
Experiment example 20, this fact appears to indicate that the basic
performance of the adhesive itself is unchanged between this
experiment and EP-106. In fact, because the adhesive in this
experiment contains elastomer and the linear expansion coefficient
ought to be close to that of metal owing to the admixture of
filler, it is expected from common knowledge in the art that good
results should be achievable after experiencing vibrations or after
experiencing elevated temperatures.
Experiment Example 24
Adhesive
[0140] The commercially available epoxy adhesive EP-106 was
acquired. In addition, the glycidyl methacrylate-ethylene copolymer
Bondfast E (available from Sumitomo Chemical, Japan), which is a
polyolefin resin, was acquired and freeze-fractured at a liquid
nitrogen temperature to obtain a powder that passes through a 30
.mu.m sieve. 100 g of the epoxy adhesive EP-106, 5 g of the above
powdered polyolefin resin and 10 g of the glass fibers RES03-TP91
were placed in a polyethylene beaker, thoroughly stirred and left
to stand for one hour, after which the contents were again stirred
to blend together the ingredients. This mixture was used as the
epoxy adhesive composition. Aside from using the resulting adhesive
composition in place of EP-106, this experiment was carried out in
exactly the same way as Experiment example 21. Two days after
curing the adhesive, the test pieces were subjected to tensile
breaking tests, whereupon the average shear strength at break for
the four pairs was 70 MPa.
Experiment Example 25
Fabrication of Commercial-Type Prepreg
[0141] Ten parts by weight of a commercially available brominated
bisphenol A-type solid epoxy resin (EPC-152, available from
Dainippon Ink & Chemicals, Japan), 13.9 parts by weight of a
bisphenol A-type liquid epoxy resin (EP-828, available from Yuka
Shell Epoxy, JApan), 15 parts by weight of tetraglycidyl
diaminodiphenylmethane (ELM-434, available from Sumitomo Chemical,
Japan), 24.8 parts by weight of a bisphenol F-type liquid epoxy
resin (EPC-830, available from Dainippon Ink & Chemicals,
Japan), 25 parts by weight of 4,4'-diaminodiphenylsulfone
(4,4'-DDS, available from Sumitomo Chemical, Japan) as curing
agents and 0.3 part by weight of BF.sub.3-monoethylamino complex
(BF3-MEA), 8 parts by weight of a weakly crosslinked
carboxyl-terminated solid acrylonitrile butadiene rubber (DN-611,
available from Nippon Zeon, Japan) and 3 parts by weight of the
thermoplastic resin hydroxyl-terminated polyether sulfone
(PES-100P, available from Mitsui Toatsu Chemicals, Japan)(the total
amount of these ingredients being 100 parts by weight) were mixed
at standard temperature and the mixture was formed into a sheet
using rollers.
[0142] The resulting resin films were set in a prepreg machine and
were joined under applied pressure from both sides of carbon fibers
(T-300, available from Toray, Japan) arranged in parallel in one
direction as reinforcing fibers by a conventional method, thereby
forming a prepreg having a resin content adjusted to 38%. The fiber
basis weight was 190 g/m.sup.2. Prepregs commercially available
from various domestic corporations appear to be manufactured by
this type of method.
Experiment Example 26
Production and Evaluation of a Composite
[0143] A5052 aluminum alloy sheet stock having a thickness of 1.6
mm was cut to form a plurality of rectangular pieces measuring
45.times.15 mm. Liquid treatment on these pieces was performed in
the same way as in Experiment example 18. That is, the pieces were,
in order, degreased with an aqueous solution of the aluminum alloy
degreaser NE-6 and rinsed with water, subjected to preliminary
pickling in a 1%-aqueous solution of hydrochloric acid and rinsed
with water, alkali etched in a 1.5%-aqueous solution of caustic
soda and rinsed with water, neutralized in a 3%-aqueous solution of
nitric acid and rinsed with water, finely etched in an aqueous
solution of hydrazine monohydrate and rinsed with water, oxidized
in an aqueous solution of hydrogen peroxide and rinsed with water
and finally dried by being placed for 15 minutes in a hot air dryer
set to 67.degree. C.
[0144] After drying, the aluminum alloy pieces were collected
together, wrapped in aluminum foil and stored. On the same day, the
aluminum alloy pieces were taken out and the liquid single-pack
dicyandiamide-curable epoxy adhesive EP-106 (available from
Cemedine, Japan) was lightly applied to the ends. The pieces were
then placed in a desiccator with the adhesive-coated sides facing
upward, the pressure was reduced to 3 mmHg with a vacuum pump and
left in that state for 1 minute, following which air was
introduced, returning the desiccator to standard pressure. This
operation of reducing the pressure and then returning to standard
pressure was repeated three times, following which the pieces were
removed from the desiccator.
[0145] Meanwhile, the firing jig 1 shown in FIG. 1 was set up, a
mold release film 17 obtained by cutting 0.05 mm-polyethylene film
into strips was laid down within the metallic mold body 2 and the
aluminum alloy sheet 11 prepared above was placed thereon. A plain
weave cloth of carbon fibers (T-300, available from Toray, Japan)
that had been separately cut was laid down as the prepreg 12 in
FIG. 1. Three layers were built up thereon while applying to the
consecutive layered surfaces an epoxy adhesive (EP-106) discharged
from a syringe. Next, a mold release polyethylene film piece 13 was
placed on the aluminum alloy sheet 11 side, following which the
size of the cloth was changed and five layers were similarly built
up while again applying adhesive.
[0146] The amount of EP-106 used was about 1 cc. PTEF blocks 14 and
15 were placed on top of the polyethylene film pieces 13 to hold
everything down and then the entire assembly was placed in a hot
air dryer. Within the dryer, a 0.5 kg iron weight was placed on top
of each PTEF block 14 and 15, following which the dryer was turned
on and the temperature was raised to 135.degree. C. Heating at
135.degree. C. was carried out for 40 minutes, after which the
temperature was raised further to 165.degree. C. over a period of 5
minutes and the temperature was held at 165.degree. C. for 20
minutes. The power was then turned off and the dryer was left to be
in cooling state with the door closed. The next day, the assembly
was taken out of the dryer, the shaped object was released from the
mold and the polyethylene film pieces 13 and 17 were peeled off,
giving the aluminum alloy composite 10 shown in FIG. 2. The same
procedure was repeated to give eight pieces of aluminum alloy
composite 10 obtained by uniting aluminum alloy with CFRP.
[0147] On the second day after joining, four pieces were subjected
to tensile break tests. The CFRP portion was sandwiched between two
pieces of 1 mm thick SUS304 stainless steel polished with sandpaper
and the resulting assembly was secured by clamping between chuck
plates. The average shear strength at break for the four pairs was
62 MPa, which was very strong. The joining surface area was
calculated as l.times.m in FIG. 2.
[0148] Next, with regard to the remaining four pieces, the united
objects were clamped in a tensile testing machine in the same way
as described above and subjected to a strain of 30 MPa, at which
point the pulling operation was stopped and the specimen was held
in this state for 10 minutes. The chuck was subsequently loosened
and the specimen was removed from the testing machine and allowed
to rest. The next day, when these specimens were subjected to a
tensile breaking test, the average shear strength at break was 60
MPa. There was thus no particular sign of a decline in the joining
strength.
Experiment Example 27
Production and Evaluation of a Composite
[0149] Using 45.times.15 mm rectangular pieces of A5052 aluminum
alloy sheet stock having a thickness of 1.6 mm like those in
Experiment example 26, similar test pieces for measuring adhesive
strength were fabricated. That is, adhesive-coated aluminum alloy
pieces were prepared by applying an adhesive to the aluminum alloy,
placing the adhesive-coated alloy pieces in a desiccator and
repeating three times the operation of pressure reduction with a
vacuum pump, followed by return to standard pressure. Next, the
metallic mold 1, 2 and 3 shown in FIG. 1 was set up, a 0.05 mm
thick polyethylene film cut into strips 4 was laid down in the mold
cavity and the aluminum alloy piece described above was placed on
top as 11. The procedure up to this point was the same as in
Experiment example 26, while the CFRP prepreg fabricated in
Experiment example 25 was used.
[0150] That is, three sheets of the prepreg in Experiment 25 that
had been cut were built up in layers, then polyethylene film 13 was
placed on the aluminum alloy side, following which five sheets of
prepreg of a different size were again built up in layers. A PTFE
weight 9 was then placed on top and the assembly was placed in a
hot air dryer. Within the dryer, a 0.5 kg iron weight was placed on
top of each PTEF block 9, following which the dryer was turned on
and the temperature was raised to 135.degree. C. Heating at
135.degree. C. was carried out for 60 minutes, after which the
temperature was raised further to 165.degree. C. over a period of
10 minutes and the temperature was held at 165.degree. C. for 60
minutes. The power was then turned off and the dryer was left to be
in cooling state with the door closed. The next day, the assembly
was taken out of the dryer, the shaped object was released from the
mold and the polyethylene film was peeled off, giving the shaped
object shown in FIG. 2.
[0151] On the second day after joining, the objects were subjected
to tensile breaking tests. The CFRP portion was sandwiched between
two pieces of 1 mm thick SUS304 stainless steel polished with
sandpaper and the resulting assembly was secured by clamping
between chuck plates. The average shear strength at break for the
four pairs was 55 MPa, which was very strong. The joining surface
area was calculated as l.times.m in FIG. 2.
* * * * *