Goal of this experiment is to gather preliminary information on the opacity of powdered TiO2 (via reflectance and or absorbance) in sorbitol propellant. Emphasis is on the percent TiO2 that can provide significant light blocking/reflectance and the propellant depth at which that light protection is provided.

Sorbitol propellant has some unusual burn characteristics some of which may be related to the propellant melting before it ignites in the motor: A hypothesized phenomena called Flushing may be one of the examples of this:
http://www.nakka-rocketry.net/knsbchar.html
If, during the burn, infrared IR light is to be prevented from penetrating the yet as unburnt grain then an opacifier could provide this protection.
The black opacifiers such as charcoal, graphite, lampblack, or carbospheres can provide this light IR blockage. However being black (IE: light absorbtive) they will capture the IR heat in the burn front layer (see note**). Also: being carbon they will also act as fuel in the burn thereby changing the burn chemistry.
If an opacifier is needed that will not only block IR penetration but will also not absorb IR heat at the burn surface then a "white" reflective compound is needed.
TiO2 can provide this IR reflective quality: see http://preview.tinyurl.com/b6pzo8

Not only this but TiO2 should not react during a sugar propellant burn but will be ejected chemically unchanged during the burn. Propep analysis and SugPro chemistry discussions suggest this is the case, Thanks Scott J. for the chemical info/discussions; they got me out of my theory chair and into the testing lab :-) (back yard lab that is ;-)

Even if not reactive TiO2 could still act as a catalyst so experimenting is needed to investigate this possibility.*

Now, for the sake of discussion, let's say that:
1. Flushing is not an issue with sorbitol propellant,
2. TiO2 does not react chemically during the burn,
3. and TiO2 does not act as a catalyst.
TiO2 could still effect burn rates based on the Summerfield theory (Aeroconsystems link: 3rd paragraph) especially under pressure, by reducing the IR heat that can be pumped from the luminous layer back into the lower layers.

There is information on the web about the use of TiO2 in AP formulas:
Controlling burn rate in AP with TiO2
Using nano-size particles in rocket propellant

TiO2 powder is a strong reflector of infrared light and visible (white) light. Because of this, preliminary testing was performed with white light for simplicity and availability of equipment.

A 100 gram batch of plain 65/35 KNO3/Sorbitol was made with screening and mixing.

The Potassium nitrate (KNO3) was fine powder (Stock #C170 - OX) from Firefox
The majority of the KNO3 crystals measured 20 microns to 200 microns
The smaller particles tended to stick to the larger ones


The Sorbitol was Sorbo-Gem food grade powder. This grade of sorbitol was available from PVC ONLY
The majority of the sorbitol particles measured 10 to 200 microns
The sorbitol will melt so crystal size is probably irrelevant to the results.


The TiO2 is very fine powder purchased at a rockshop. It is used by rockhounds as a rock polish.
The image below is at 400x and each line unit represents 2.2 microns.
Particles range from 0.5 microns to 2 microns. The objects that look larger are clumps of particles.

The fine particle size is for the purpose of giving rocks a smooth polished surface but this size is also the most efficient for reflecting light in the visible and infrared spectrum .

(500nm to 2,000nm particle size equates to the wavelengths of blue light up to the wavelengths of near infrared.)

5 and 10 gram aliquots were taken from the initial 100 gram batch and TiO2 was added so that samples containing 10%, 5%, 2.5%, 1%, 0.5%, and 0% TiO2 were made.
Dry melting was performed with a wax double boiler.

10 gram samples melting:

5 gram samples melting:


Specimens were pressed between microscope slides with spacers to make cured samples that were 0.15mm thick (150 microns) and another set that were 1mm thick (1000 microns)


TiO2 powder is a strong reflector of infrared light and white light. Preliminary testing was performed with white light for simplicity and availablity of equipment.

Three methods for imaging:

Image Method 1.
Under microscope ~40x 0.15mm (150 microns) thick:
Plain 65/35 sorbitol propellant dry melt method:
Top half of image is the propellant bottom half is clear glass to show light intensity applied to sample:

This represents transmitted light
IE: light that is penetrating the propellant from the side opposite the viewer.
One can see the light that has passed through the crystals and micro-bubbles of the propellant.


Now the same sampling process with 2.5% added TiO2:
Under microscope ~40x 0.15mm (150 microns) thick:
2.5% TiO2 added to 65/35 sorbitol propellant dry melt method:

As the reference would suggest TiO2 has amazing opacity:

This is not due to light absorbance, otherwise the surface of TiO2 would look black or colored.
This opacity (light blockage) is due to TiO2 extreme reflectance/scattering of the light at it's surface.
Even at the 2.5% level almost complete light blockage was seen with webthickness as thin as 150 microns:

To allow higher applied light intensity all the 150 micron thick samples were photographed at 100x with the objective focused in the center area of the samples.
Areas were chosen to demonstrate embedded crystals and air bubbles as well as the TiO2 opacity.

WOW! Those bubbles make a good point for vaccum degassing:



At 10% details were not visible (above right). The rheostat was maxed out, so I cranked up the light intensity by opening the light condensor diaphragm even further: (below).




The microscope imaging has the advantage that the cameras can quantitatively meter the light that has penetrated a sample.
The following tables and graphs shows the shutter settings from a film SLR camera's light meter and the shutter times from the digital camera:

Magnification 40x - SLR camera set to 3200 ASA - 150 micron thick samples - %TiO2 versus resulting Shutter speed:

Thickness	%TiO2	Shutter speed	ASA	Milliseconds
0.15mm	No Sample	1/1000 sec	3200	1
0.15mm	0%	1/500 sec	3200	2
0.15mm	0.5%	1/250 sec	3200	4
0.15mm	1%	1/125 sec	3200	8
0.15mm	2.5%	1/60 sec	3200	17
0.15mm	5%	1/30 sec	3200	33
0.15mm	10%	1/30 sec	3200	33

Graphing this data for %TiO2 vs Exposure in milliseconds:



It would be expected that the thicker samples will block more of the light from penetrating the propellant as can be seen in the following table and graph:
Magnification 40x - SLR camera set to 3200 ASA - 1mm thick samples - %TiO2 versus resulting Shutter speed:

Thickness	% TiO2	Shutter speed	ASA	Milliseconds
1 mm	No Sample	1/1000 sec	3200	1
1 mm	0%	1/125 sec	3200	8
1 mm	0.5%	1/30 sec	3200	33
1 mm	1%	1/15 sec	3200	67
1 mm	2.5%	1/8 sec	3200	125
1 mm	5%	1/4 sec	3200	250
1 mm	10%	1/4 sec	3200	250

This data graphed along with the data from the thinner samples:


This light metering was repeated with the 150 micron samples using a digital camera at a magnification of 100x:
Results as follows:

Thickness	% TiO2	Exposure in Milliseconds
0.15mm	0%	4.1
0.15mm	0.5%	13.9
0.15mm	1%	17.4
0.15mm	2.5%	26
0.15mm	5%	31
0.15mm	10%	692 Wow! That is a jump not seen with the film camera metering

And this data graphed:


The SLR exposure data versus the Digital exposure data for the thin prep (150 micron) samples:




Image method 2.
So what did these samples look like to the naked eye ?
Here they are over an Osram Diastar 200 slide viewer along with some thicker samples (1mm = 1000 microns thick):
Macroscopic view: Each specimen ~ 1/2 inch across. Specimen depth listed in photo:
White Light transmission/vs blockage:



Image Method 3.
So what about surface reflectance?
Macroscopic view: Each specimen ~ 1/2 inch across. Specimen depth listed in photo:
White Light surface reflectance:


All of this is irrelevant if the additive is detrimental to the burn.
To get a preliminary idea of the effect of TiO2 I did 1 atm strand burn tests:

TiO2 in sorbitol propellant:

% TiO2 added	1 atm burn rate	notes
0%	8.5 seconds/inch	normal burn
0.5%	8.5 seconds/inch	normal burn
1%	8.5 seconds/inch	normal burn
2.5%	~8 seconds/inch	normal burn
5%	~8 seconds/inch	normal burn
10%	10 seconds/inch	subdued burn and molten "beads" forming

I did not get a video of the 10% strand burn but did make this "artists color pallette" that I caught on video during burn:
The video is kinda blurry and dizzy so this picture helps get you oriented before watching the video:

The video of the propellants burning in 1 atm on YouTube


reduced size/quality of video download TiO2.wmv (File size 250 KB sized down for easier dial up.)

Conclusions

As was discussed on SugPro a white reflecting opacifier might be a solution in search of a problem.
However: Should the need for this type of opacifier be found this experiment should provide data for where to start with the concentration of TiO2 and the approximate depth at the burn surface that is being IR protected.

It is apparent that visible light is blocked from penetrating sorbitol propellant by TiO2 reflecting the light back towards the light source. This occurs to some extent even at 0.5% to the depth of only 150 microns.
What this might mean for effect on burn rates at various pressures is something that would need motor thrust tests or strand tests under increased pressure.

Some 2006(AR)Email discussions of this opacifier with Fred Azinger, a HPR AP EX rocketeer, brought up some statements regarding AP that helped me to succintly see TiO2's possible benefit for sorbitol propellant: "TiO2 is interesting since it creates a plateau in the burn rate -vs- pressure curve to give a nice self-limiting effect."
"There are many patents written about the function of TiO2 for burn rate retardant. Basically, it works the same as its sun-block effect....it shields the propellant from IR heating and thus slows the liquidification of the surface, slowing the burn."
I need to reread McCreary's book and Nakka's site on "a" and "n" values to see if I can understand where these concepts fit in with TiO2's potential effect.

Another item from SugPro was a name for this propellant. Ken called it "KNSB Ultra-White". ...I Like It!

I had wanted to measure transmission and reflectance of IR light, but was convinced by the reference that TiO2 responded similarly with IR and White light.

To reiterate a point in the introduction: Even if not reactive in the burn TiO2 could still act as a catalyst so experimenting is needed to investigate this possibility.*

Side notes and observations:

* 2007 Email discussion with B.C.

** Note: There are black opacifiers that are IR reflective but that is beyond the scope of this project.

Case/Liner Bonding ??
When melting the various % formulas I noticed that the propellants with TiO2 seemed harder to scrape off of the teflon pan surface. This may just be an illusion since the whiter propellants are more easily seen even in thin preps. But this still caused me to wonder if upon cooling that they might bond better.
Of course I don't know anyone who casts into teflon liners or cases. LOL
But I do wonder how it might bond to other material such as phenolic or aluminum or paper.
(so many questions so little time :-)

Melt Viscosity ?
When repeating the experiment with the 0%, 0.5%, and 1.0% samples I did not see any effect on melt viscosity.
Image below shows height to which molten propellant could be pulled to a "peak".



Reference Links:

Richard Nakka's sorbitol propellant site

Burn Characteristics of Sorbitol Based Propellants By Chuck Knight @ http://www.nakka-rocketry.net/knsbchar.html

Introduction to IR-Reflective Pigments By Mark Ryan: Paint and Coatings Industry site.

Nonsteady Burning and Combustion Stability of Solid Propellants By Luigi De Luca, page 256

The dP/dt Failure in Propellants and Ignition Compositions: William Colburn, Aeroconsystems site

Back to the 24mm Sorbitol Drysophila Project page

::: Made with CoffeeCup : Web Design Software & Website Hosting :::