Properly bonding composites and what your government doesn't want you to know.

The Rocketry Forum

Help Support The Rocketry Forum:

This site may earn a commission from merchant affiliate links, including eBay, Amazon, and others.
Yes, I agree that the statement is incorrect and/or confusing. Presumably he meant sharing of electrons (in the "outer shell"), which is characteristic of chemical bonds.

However, I'm still not convinced that epoxy makes a chemical bond with fiberglass (silica). It might, but it needn't; the epoxy creates a matrix in which the fibers are embedded. Any chemists here able to clarify?

And in general, I think the original post has a lot of value. I've been sanding with finer grit and cleaning the surfaces before sanding also as a result of reading it. Things I've learned here are reflected in my recent Epoxy Basics video.


You guys are correct in that I was incorrect in my post. You are not forming a subatomic bond. You are increasing the surface activation of the part causing the the wetting to increase.

https://en.wikipedia.org/wiki/Wetting

Wetting is the ability of a liquid to maintain contact with a solid surface, resulting from intermolecular interactions when the two are brought together. The degree of wetting (wettability) is determined by a force balance between adhesive and cohesive forces. Wetting deals with the three phases of materials: gas, liquid and solid. It is now a center of attention in nanotechnology and nano-science studies due to the advent of many nano-materials in the past 2 decades (e.g. graphene,[1] carbon nano-tube).

Wetting is important in the bonding or adherence of two materials. Wetting and the surface forces that control wetting are also responsible for other related effects, including so-called capillary effects. Regardless of the amount of wetting, the shape of a liquid drop on a rigid surface is roughly a truncated sphere. Various degrees of wetting are summarized in this article.
 
peel ply finish does not have a better bonding surface than a recently sanded or blasted surface. Peel ply is generally used where you will be doing additional layups of laminate and its not worth the time or expense to sand or blast. Some of the articles I linked earlier have DOD research on different bond preps and comparative strengths.

silane is I believe used as what I have always heard called as a "sizing"

Raw carbon fibers do come with a "sizing" to improve the bonding or the resin matrix inter laminerly as you sand to prep you will be removing this layer on the outside of the fibers.
 
peel ply finish does not have a better bonding surface than a recently sanded or blasted surface. Peel ply is generally used where you will be doing additional layups of laminate and its not worth the time or expense to sand or blast. Some of the articles I linked earlier have DOD research on different bond preps and comparative strengths.

silane is I believe used as what I have always heard called as a "sizing"

Raw carbon fibers do come with a "sizing" to improve the bonding or the resin matrix inter laminerly as you sand to prep you will be removing this layer on the outside of the fibers.

I don't mean to challenge your expertise or experience-please don't take this as hostile.
I believe Bob Krech posted (or maybe sent directly to me) a paper written at McDonnell Douglas, which concluded that dacron peel ply created a better bonding surface than grit blasting; I believe this was because gritblasting damaged the underlying fibers. I imagine it depends on how the surface is sand-blasted and cleaned. It also found that sanded or ground surfaces were worse, because of damage to the fibers, at least I think? I believe that this is also the source CarVac is referring to.

I believe the context of the study was for adding more laminations of fabric? Is there a reason structural bonding would be different than adding more laminations?

Again, I bring this up not because I don't believe you or want to argue, but because I am confused and want to understand more. PM me if you would rather explain that way.
 
check out page 31 of the first link I posted in background reading. While peel ply does leave a rough finish the resin surface does not fracture that well.
 
check out page 31 of the first link I posted in background reading. While peel ply does leave a rough finish the resin surface does not fracture that well.

Thank you for pointing this out, it will drastically change how I build.
 
flynfrog;653048... Raw carbon fibers do come with a "sizing" to improve the bonding or the resin matrix inter laminerly ...[/QUOTE said:
That is news to me! Some carbon fabric has a sizing to help the fabric maintain structure before lamination. Most do not. The ones which do tend to be spread tow fabrics. The sizing is a light adhesive layer or epoxy resin coat which keeps the fibers lightly stuck to each other to aid in handling. There is also unicarbon fabric which utilizes tracer threads to keep the fibers under control, and in some cases these threads are actually very thin beads of hot glue. Perhaps this sort of thing is what you are referring to? If not, I'd love to see an article on it!

Here is something I found on the subject: https://strong.groups.et.byu.net/pages/articles/articles/surface.pdf - but it indicates the purpose of the sizing is not to improve the bond but to reduce fiber breakage during manufacturing. It is not the same sizing to which I was referring.

Thanks,
Gerald
 
The sizing ist mostly out of my realm The little I know is to get a sizing that is compatible with you resin system.

https://www.hexcel.com/Products/CF_ContSizing


Another article I found.
Surface treatment and sizing
The next step is critical to fiber performance and, apart from the precursor, it most differentiates one supplier’s product from its competitors’ product. Adhesion between matrix resin and carbon fiber is crucial in a reinforced composite; during the manufacture of carbon fiber, surface treatment is performed to enhance this adhesion. Producers use different treatments, but a common method involves pulling the fiber through an electrochemical or electrolytic bath that contains solutions, such as sodium hypochlorite or nitric acid. These materials etch or roughen the surface of each filament, which increases the surface area available for interfacial fiber/matrix bonding and adds reactive chemical groups, such as carboxylic acids.

Next, a highly proprietary coating, called sizing, is applied. At 0.5 to 5 percent of the weight of the carbon fiber, sizing protects the carbon fiber during handling and processing (e.g., weaving) into intermediate forms, such as dry fabric and prepreg. Sizing also holds filaments together in individual tows to reduce fuzz, improve processability and increase interfacial shear strength between the fiber and matrix resin. Carbon fiber producers increasingly use a sizing appropriate to the customer’s end use (see sidebar, below and “Sizing and surface treatment: The key to carbon fiber’s future?” in “Learn More,” at right). At Grafil, Carmichael adds, “we can customize surface treatment and sizing to a particular customer’s resin characteristics, as well as specific properties desired in the composite.”

https://www.compositesworld.com/articles/the-making-of-carbon-fiber


With some RTM shots we have to be careful not to over cook the carbon to risk damaging the sizing. Good article thanks.
 
Here is my help in clarifying some of the great points that guys have brought up in this thread. I would leave the sub-atomic or sub-orbital discussion for the quantum chemistry forum! Focusing on the “functional groups” might be more helpful. All materials, or rather compounds, are made up from the 115 different elements on the periodical table, these being oxygen, hydrogen, iron, aluminum, and etc. These elements based on their outer orbital configuration, once again going to skip this, will want to combine with other elements to form a more stable “functional group”. For example, one oxygen is actually not stable, so when it comes in contact with another oxygen, it bonds to it making a different oxygen to oxygen compound. This new oxygen bonded to oxygen (known as O2) compound is more stable and is really what we all require when breathing air into our lungs. Now here is another example, oxygen can also bond to two hydrogens, once again to make a more stable compound and this hydrogen-oxygen-hydrogen compound is known as water (H2O). Thus, the outer orbital configuration plays a role into what elements can bond with other elements, but my focus is the “functional group” which is made from all these different combinations of elements. Example of a “functional group”, if I take my water compound (hydrogen-oxygen-hydrogen) and replace one of the hydrogens with another compound (as long as it is stable) I can then make other compounds that have this oxygen-hydrogen group and thus, this oxygen-hydrogen group becomes a “functional group”. And now, this new compound with the oxygen-hydrogen “functional group” can mix with other materials with this same “functional group” such as water. These examples show compounds that are made from the elements that are connected by covalent bonds, covalent bonds are strong because they form and create a more stable compound. However, there is other types of bonds as well, ionic, van der wall, and hydrogen bonding. These exist and play an important part in creating compounds but do not form as strong of a bond as the covalent bond. Take a look at hydrogen bonding, this exists when the oxygen-hydrogen “functional group” interacts with another oxygen-hydrogen “functional group”, the two hydrogens create a bond, again not as strong as the covalent bond but still it is termed a bond. Here is an example of hydrogen bonding, when you fill a coffee mug full of water, you can slowly fill it until the water actually is above the rim and the water can rise above the rim and not spill out, this is because hydrogen bonding is occurring between each water molecule holding them together. Hydrogen bonding basically occurs due to the similarity of certain “functional groups” within compounds that are covalently bonded from the various elements.
This is where the “Water Break” test becomes helpful when bonding materials, when water spreads out on your bonding surface, it means that the water is mixing with the surface compounds (otherwise known as wetting out the surface). If the water beads up, then the water is not compatible with the bonding surface just like the oil and water story, they just don’t mix. But then, we all know that we can get oil and water to mix if we add some soap. Soap is made up of two “functional groups”, one of the “functional groups” mixes with water and the other “functional group” mixes with the oil, thus it is termed bifunctional. Industry will use emulsifier and coupling agent as terms to describe this bifunctional configuration. Now we can bring in “sizing”, fiberglass requires some kind of sizing in order to be fully wet out by the resin. Untreated fiberglass does not have the correct “functional group” on its surface to allow complete wet out from epoxy resins. Therefore, fiberglass is treated with a coupling agent (again, this is similar to your soap, bifunctional) and the most common one used is A1100. When purchasing fiberglass for epoxy, phenolic, or unsaturated polyester resin applications, A1100 is what you would want. A1100 is an amino-silane, the amino “functional group” will provide compatibility with the resin and the silane “functional group” is compatible with the fiberglass surface.
Now to the epoxy resins, when epoxies cure (this is known as crosslinking) some of the covalent bonds in the epoxy resin are broken and are then re-formed with the hardener creating new “functional groups”. This new “functional group” is typically an oxygen-hydrogen “functional group” and this group is very good at forming the hydrogen bond, like the water in the coffee mug example. This is the bond that makes epoxies so good as an adhesive. For example, a properly prepared piece of aluminum for an aircraft application will have a good uniform oxygen-hydrogen “functional group” on the aluminum surface and when an epoxy adhesive is properly used to bond two pieces of this aircraft aluminum, the oxygen-hydrogen “functional group” from the aluminum and the oxygen-hydrogen “functional group” from the epoxy will form this hydrogen bonding network. The same thing occurs with the A1100 sized/coated fiberglass. Thus, epoxies make good adhesives as long as their functionality is similar to the bonding surface functionality.
 
:eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop::eyepop:
I just lost about 3/4 of my brain cells attempting to interpret this...
Here is my help in clarifying some of the great points that guys have brought up in this thread. I would leave the sub-atomic or sub-orbital discussion for the quantum chemistry forum! Focusing on the “functional groups” might be more helpful. All materials, or rather compounds, are made up from the 115 different elements on the periodical table, these being oxygen, hydrogen, iron, aluminum, and etc. These elements based on their outer orbital configuration, once again going to skip this, will want to combine with other elements to form a more stable “functional group”. For example, one oxygen is actually not stable, so when it comes in contact with another oxygen, it bonds to it making a different oxygen to oxygen compound. This new oxygen bonded to oxygen (known as O2) compound is more stable and is really what we all require when breathing air into our lungs. Now here is another example, oxygen can also bond to two hydrogens, once again to make a more stable compound and this hydrogen-oxygen-hydrogen compound is known as water (H2O). Thus, the outer orbital configuration plays a role into what elements can bond with other elements, but my focus is the “functional group” which is made from all these different combinations of elements. Example of a “functional group”, if I take my water compound (hydrogen-oxygen-hydrogen) and replace one of the hydrogens with another compound (as long as it is stable) I can then make other compounds that have this oxygen-hydrogen group and thus, this oxygen-hydrogen group becomes a “functional group”. And now, this new compound with the oxygen-hydrogen “functional group” can mix with other materials with this same “functional group” such as water. These examples show compounds that are made from the elements that are connected by covalent bonds, covalent bonds are strong because they form and create a more stable compound. However, there is other types of bonds as well, ionic, van der wall, and hydrogen bonding. These exist and play an important part in creating compounds but do not form as strong of a bond as the covalent bond. Take a look at hydrogen bonding, this exists when the oxygen-hydrogen “functional group” interacts with another oxygen-hydrogen “functional group”, the two hydrogens create a bond, again not as strong as the covalent bond but still it is termed a bond. Here is an example of hydrogen bonding, when you fill a coffee mug full of water, you can slowly fill it until the water actually is above the rim and the water can rise above the rim and not spill out, this is because hydrogen bonding is occurring between each water molecule holding them together. Hydrogen bonding basically occurs due to the similarity of certain “functional groups” within compounds that are covalently bonded from the various elements.
This is where the “Water Break” test becomes helpful when bonding materials, when water spreads out on your bonding surface, it means that the water is mixing with the surface compounds (otherwise known as wetting out the surface). If the water beads up, then the water is not compatible with the bonding surface just like the oil and water story, they just don’t mix. But then, we all know that we can get oil and water to mix if we add some soap. Soap is made up of two “functional groups”, one of the “functional groups” mixes with water and the other “functional group” mixes with the oil, thus it is termed bifunctional. Industry will use emulsifier and coupling agent as terms to describe this bifunctional configuration. Now we can bring in “sizing”, fiberglass requires some kind of sizing in order to be fully wet out by the resin. Untreated fiberglass does not have the correct “functional group” on its surface to allow complete wet out from epoxy resins. Therefore, fiberglass is treated with a coupling agent (again, this is similar to your soap, bifunctional) and the most common one used is A1100. When purchasing fiberglass for epoxy, phenolic, or unsaturated polyester resin applications, A1100 is what you would want. A1100 is an amino-silane, the amino “functional group” will provide compatibility with the resin and the silane “functional group” is compatible with the fiberglass surface.
Now to the epoxy resins, when epoxies cure (this is known as crosslinking) some of the covalent bonds in the epoxy resin are broken and are then re-formed with the hardener creating new “functional groups”. This new “functional group” is typically an oxygen-hydrogen “functional group” and this group is very good at forming the hydrogen bond, like the water in the coffee mug example. This is the bond that makes epoxies so good as an adhesive. For example, a properly prepared piece of aluminum for an aircraft application will have a good uniform oxygen-hydrogen “functional group” on the aluminum surface and when an epoxy adhesive is properly used to bond two pieces of this aircraft aluminum, the oxygen-hydrogen “functional group” from the aluminum and the oxygen-hydrogen “functional group” from the epoxy will form this hydrogen bonding network. The same thing occurs with the A1100 sized/coated fiberglass. Thus, epoxies make good adhesives as long as their functionality is similar to the bonding surface functionality.
 
Thank you Hy test for clarifying. Any chance you can condense it down to a few lines I can add to the first post. I totally geeked out on your explanation but it might make some peoples eyes glaze over in its entirety.
 
Here's my attempt to condense the key part:
"Untreated fiberglass does not have the correct chemical configuration on its surface to allow complete wet out from epoxy resins. Therefore, fiberglass is treated with a coupling agent and the most common one used is A1100. When purchasing fiberglass for epoxy, phenolic, or unsaturated polyester resin applications, A1100 is what you would want."

Thanks Hytest for the great explanation! (This also answers my question about chemical bonding: yes it occurs, at least with "sized" flberglass.)
 
https://en.wikipedia.org/wiki/Adhesion explains how 2 items can be joined together by an adhesive.

Many adhesives, glues, paints that the consumer purchases don't adhere well to one or both surfaces of the materials being joined or coated. That's because the stores don't carry the professional "primers", "surfactants", "emulsifiers" or "bonding agents" that are necessary to make the products strongly adhere to the surfaces being joined or coated. The "pros" buy their supplies from distributors that carry a complete product line for one or more manufacturers where as hobbyists are likely to purchase their supplies from a "big box" store or crafts shop that don't carry an entire product line, for example, lots of glues, adhesives and paints, but no or limited surface primers, with untrained sales staff without manufactures datasheets or application notes.

Examples: primers for paints, primers and bonding agents for adhesives, fluxes for solder, seizing for wall paper, etc.

Adhesives should only be applied to cleaned surfaces. No adhesives will stick well to a dirty and/or oily surface.

RTV Silicone adhesives will not adhere well to metals and many other materials without pretreating the surfaces with a primer.

Paints will not adhere well to many surfaces without a primer, or over rusted metal with an surface oxide converter.

Fluxes clean and prevent metal surfaces so that the solder can alloy with the surface to make a solid bond. Without flux, solder will not wet and adhere to the metal and will at best encapsulate it making a "cold solder joint" that is easily broken by movement or vibration.

Any adhesive or glue will not adhere to polyethylene, polypropylene, Teflon or similar smooth plastics without an etching "surface activator". PVC and CPVC cement will not form a strong bond without a primer.

Wall paper will not adhere well to a plaster wall without seizing.

As mentioned most epoxy will not adhere well to fiberglass or carbon fiber without seizing. Fortunately most commercial cloths have seizing.

Minimal APCP propellants contains AP and a binder, typically 88% HTPB and 12% PAPI which forms a rubber. The bonding of the rubber to the AP crystals is greatly enhanced by ~percent level addition of surfactants, bonding agents. and gas releasing agents (https://en.wikipedia.org/wiki/Simethicone) that increase the adhesion of the rubber to the AP. The addition of metal fuels may also require addition of antioxidant compound to prevent the oxidation of the aluminum until the motor is ignited. Antioxidants also prevent the rubber from cracking or becoming brittle with age.

These are some "tricks of the trade" that separate the hobbyists from the pros. The hobbyist can often obtain this knowledge by downloading and reading the Tech sheets and Application Notes from the manufacturers websites.

FWIT

Bob
 
I've been using LORD® 310 adhesive, a modified, thixotropic, two-component epoxy adhesive system for several years to make CF structures. What's nice about it is that it can be used to over 400 F (204 C) with room temperature cure. It can be used to bond many types of prepared metals, prepared rubber, urethane and plastics. Originally formulated for primerless adhesion to SMC, LORD 310 adhesive is used to bond disk drives, down-hole oil field equipment, automotive body panels and spoilers, and vibration dampening mounts. LORD 310 adhesive can be either room temperature cured or heat cured for faster processing.

I order it from MMC in dispenser tubes and use a manual gun and a static mixer tube to apply it directly to the parts. The components are white and black any you can see that it is properly mixed in the static mixer tube as it becomes gray. It is not the least expensive epoxy, however if you need a structural epoxy for fin attachment or for motor mounts and bulkheads, it tough to beat IMO. If you use the gun and the static mixer applicator, you epoxy is always optimally mixed.

Being a professional epoxy, the information on the linked webpage is really useful as it describes what bonding agent and primers are available for difficult bonds and the product selector booklet, technical data sheet, and bond design guide are top notch professional documents.

Bob
 
@flynfrog

I wanted to know if its benefical to use compressed air before you apply epoxy to a surface? I have been doing this for awhile and wanted to see if it worth the effort.

So I will sand the area, wipe it clean with alcohol then I spray it was a can of compressed air trying to remove any additional dust. Finally, I apply the epoxy, is this worth it? Or am I being to cautious?
 
There probably isn't much dust on the surface after you wipe. Keep in mind canned air isn't really air and air out of a compressor is full of oil and water unless you have an expensive filter.
 
I've enjoyed this thread a lot as I prepare to do my first full-composite build, but I have a question about edge-to-face bonding, like we most commonly do, as opposed to the face-to-face bonding showed in the excellent write up. With the exception of gluing a coupler in to a nose or an airframe tube, the bulk of our bonds are either the edge of a CR against a "face" of the MMT or airframe, or the "face" of a fin against the edge of the airframe in TTW applications (and similarly to the CR, the edge of a fin against the face of the MMT). The act of cutting the ring, fin or fin slot breaks the fibers, so what, if any, additional treatment should be applied to edges? After reading the article I know how to prep the tubes, and imagine all that is required on the edges is a light sanding to flush the edge, a solvent wipe until clean, and then dry, right?
 
To clarify beyond a shadow of a doubt: Denatured alcohol or acetone are the recommended cleaning solvents for glass parts prior to bonding?
 
soopirV: If I am understanding correctly you are asking about the cut edge where the fin goes. A scuff is needed to reactivate the surface and a clean like you posted. Keep in mind you want to prep the surface where you fillets will be. Once a crack has started from an area of poor prep it will only grow even if the rest of the bond is good.

Yes Old dude I recommend acetone followed by alcohol. MEK is also acceptable.
 
I see the value in clamping horizontal connections, but this may not be practical for 90 degree ones. What, if any, steps should be taken for those connections? I can imagine all sorts of elaborate rigs for pressing fins to tubes, but doubt many would engage in the process, and I wonder how much the fillet makes up for that.
 
I see the value in clamping horizontal connections, but this may not be practical for 90 degree ones. What, if any, steps should be taken for those connections? I can imagine all sorts of elaborate rigs for pressing fins to tubes, but doubt many would engage in the process, and I wonder how much the fillet makes up for that.

You have to be a bit careful attaching fins that you don't squeeze all of the epoxy, JB Weld, whatever, out from under the root with too much pressure. If you have through the wall mounted fins, the root attachment is less critical. The root attachment of surface mounted fins is important.
 
I see the value in clamping horizontal connections, but this may not be practical for 90 degree ones. What, if any, steps should be taken for those connections? I can imagine all sorts of elaborate rigs for pressing fins to tubes, but doubt many would engage in the process, and I wonder how much the fillet makes up for that.

The main point of the clamping to push out air and make sure your bond is free of voids. You don't need to be super tight but you do need to remain in position until handling strength is reached. In the case of a MD rocket your fillets are the critical part of the bond. When I do mine I tack them with CA and build nice fillets.
 
To clarify beyond a shadow of a doubt: Denatured alcohol or acetone are the recommended cleaning solvents for glass parts prior to bonding?

As long as the cleaning solvent is non-aqueous; i.e. no water in the solvent container. This may be problematic. How do you fully dry or extract all the water from a solvent which has a natural affinity for drawing water out of the atmosphere.
 
We were trained through lots and lots of research n development to pull peel ply, sand forty grit, remove and peel ply fibers, solvent wipe clean rag, WET OUT surface to be bonded, PYRAMID bonding epoxy, set bond until squeeze out. Clean excess. Pyramid helps eliminate air bubbles. ( voids )

If you get a noticeable void after curing (only seen in fiberglass obviously unless tap testing or sonogram )you can drill a small pin hole into the void and fill it.
Lots of fillers can be added for various fillets and bonds to add to consistency and strength.

Not sure if any of this was mention before, but coming from a company that deals with strictly FG and CF it's a pretty solid tip.

BtW, if clamping, I wouldn't clamp super tight. I'd make sure I had a predetermined bond line thickness even across the entire surface.
 
Last edited:
@flynfrog

I wanted to know if its benefical to use compressed air before you apply epoxy to a surface? I have been doing this for awhile and wanted to see if it worth the effort.

So I will sand the area, wipe it clean with alcohol then I spray it was a can of compressed air trying to remove any additional dust. Finally, I apply the epoxy, is this worth it? Or am I being to cautious?

Compressed air will only add unwanted foreign "release" oils. DO NOT USE COMPRESSED AIR out of a can. Vaccuum only.
 
Rule 3: Learn your glue not all epoxies are equal. They all have different properties learn what these mean and how they effect your bond.

This is a really important rule!!! Most rocket parts fail from toughness or bond prep and not shear strength (really)

One other thing is double cup / and weigh if you hand mix.

Mike (been doing composites for a really long time) K
 
Is wet sanding acceptable for fiberglass surface prep for epoxying? It would greatly reduce the amount of airborne dust. I wear a good respirator but don't have a dedicated work shop at home so I'm worried about fiberglass dust in the air. I go outside to cut fiberglass, but having to do so for surface prep would be a PITA (New England winter = sunset at 430pm, snow, and freezing temperatures).
 
Back
Top