The Capillary Action of Gasoline: Debunking the Gravity-based Siphon Theory

[BadBrain]

My Siphon Challenge FINALLY!

I've thought over my remarks concerning the theory by Dr. Stephen Hughes, Lecturer in Physics at Queensland University of Technology, concerning gravity, rather than differential air pressure, as being the primary source of motive energy for a siphon system. (See: http://www.guardian.co.uk/science/blog/2010/may/10/dictionary-definition-siphon-wrong" :

Notes & Theories science blog

Dictionary definition of 'siphon' has been wrong for nearly a century

A schoolboy error in the Oxford English Dictionary's definition of 'siphon' has come to light after nearly 100 years in print


Dictionary definition of 'siphon' has been wrong for nearly a century

A schoolboy error in the Oxford English Dictionary's definition of 'siphon' has come to light after nearly 100 years in print


The Oxford English Dictionary gives the definition of siphon as a tube for 'drawing off liquids by means of atmospheric pressure'. Photograph: Roger Tooth/Guardian

Perpetuated by dictionaries for nearly a century, it's surely the most persistent scientific howler in the history of the English language. Siphons – those ingenious plastic tubes we use to fill or drain everything from aquariums to petrol tanks – move liquid by "the force of atmospheric pressure".

Except, how could a siphon possibly work by a difference in pressure when atmospheric pressure is the same for the liquid at both ends of the tube? Bleeding obvious when you think about it. Even I can figure that out 25 years after I scraped through A level physics.

And yet according to the Guardian science desk's own coffee-stained Collins, a siphon is "a tube placed with one end at a certain level in a vessel of liquid and the other end outside the vessel below this level, so that atmospheric pressure forces the liquid through the tube and out of the vessel".

***

I've concluded that I was totally wrong, but that Dr. Hughes was only partially right.

Believe it or not, my experiences with juice boxes led me to this conclusion.

I've noticed that once I've stuck a straw into a juice box, the juice flows out spontaneously, and my only explanation for this observation was that, as I live at sea level, the juice boxes must have been filled at a higher altitude (with lower ambient air pressure) than my own home. Then I realized that what I was actually observing here was the massive power of the capillary action of water.

I then realized that, when siphoning gasoline, the initial suction at the end of the longer length of the tube (that being the end of the tube aimed into your jerry can, while the end of the shorter length is immersed into the gas tank of the car you're siphoning) is required to create an air pressure differential sufficient to get the flow going, but neither that suction (as the atmospheric pressure at the higher level of gasoline in the car's tank is actually a bit lower than the pressure pressing down upon the surface of the gasoline in your jerry can) nor gravity (as it cannot explain the tendency of the gasoline in the car's tank to rise up (i.e., AGAINST the force of gravity) into the shorter length of the hose) can explain the continuous flow of the gasoline through the hose.

Only capillary action can explain that.

And only the fact that the surface tension/capillary action of gasoline is far weaker than that of water (combined with the possibility that my juice boxes may have been filled at an altitude higher than my own (which, seeing as I live at sea level, is hardly implausible)) explains the differential results between my juice boxes and a teenager trying to earn favor with the mob, in that ambient air pressure is sufficient to create a continuous flow in the former, whereas some degree of negative atmospheric pressure (i.e., suction) is required to create a similar flow with respect to gasoline.

Gravity is only sufficient to maintain the flow by pulling down on the mass of liquid within the longer length of hose emptying into your jerry can if the volume of liquid within that longer length of hose can pull along the upward-flowing liquid behind it in the shorter length of hose due to hydrogen bonding (i.e., surface tension/capillary action). Otherwise, the upward-bound liquid within the shorter length of tube would simply fall back out of the tube back into the tank, interrupting the flow.

Ethanol exhibits no hydrogen bonding capillary action, and, therefor, should be impossible to siphon, should my hypothesis be correct. I'll try to get my hands on some pure EtOH in the near future, in order to test my hypothesis (and to celebrate my results if I'm right, and to drown my tears if I'm wrong).

***

Here's my source as to capillary action in gasoline versus that of EtOH:

Susan E. Powers (Clarkson University, Potsdam, NY), Ph.D., P.E.:

Transport and dissolution of ethanol and ethanol-blended gasoline in the subsurface

Workshop on Ethynol and Alkylates in California Automotive Fuels (April 11, 2001)

"www-erd.llnl.gov/ethanol/proceed/etohf_t.pdf"[/URL]

Please scroll down to Page 17:

"Ethanol does not spread by capillary action.

Gasoline continues to move in advance of ethanol front.

Capillary fringe is depressed."

 

[HallsofIvy]

That's not my experience. My experience with "juice boxes" is that if you put the straw in while holding the box with your fingers on the largest size, juice flows out, but not if you are holding the box by the smaller size. My conclusion is that my fingers are applying enough pressure to the box to cause the juice to come out.

 

[sophiecentaur]

I've thought over my remarks concerning the theory by Dr. Stephen Hughes, Lecturer in Physics at Queensland University of Technology, concerning gravity, rather than differential air pressure, as being the primary source of motive energy for a siphon system. I've concluded that I was totally wrong, but that Dr. Hughes was only partially right.

OF course it's not due to differential air pressure. It's due to the total difference in pressure and will depend much more on the density of liquid in the syphon. There is a significant difference in hydrostatic pressure caused by the difference in liquid column height. The difference in air pressure is actually in the WRONG direction! Air pressure is higher at the bottom than at the top, which would tend to be pushing liquid UP the tube and not down.

Your juice boxes may well have been filled under conditions of slightly positive pressure inside the machine - to keep the bugs out. But I think that they are, in fact totally filled with liquid so there is no air inside to expand. You would have to examine the sides of the boxes and see just how flat they are and whether they 'sag' when left to themselves. This would cause excess pressure inside due to the fact that the top level would have dropped by a mm or two.

In any case, they are filled in a holder with flat sides and resting on a flat surface. I reckon that, when you pick a box up to open it, the pressure of your hand around it (at the contact points with your hand) may well be higher than AP. Can you be sure that you are not just 'too rough' to guarantee no extra force on the box?

 

[BadBrain]

That's not my experience. My experience with "juice boxes" is that if you put the straw in while holding the box with your fingers on the largest size, juice flows out, but not if you are holding the box by the smaller size. My conclusion is that my fingers are applying enough pressure to the box to cause the juice to come out.

How about when you're not holding the box at all, which is the experience to which I here refer?

 

[BadBrain]

OF course it's not due to differential air pressure. It's due to the total difference in pressure and will depend much more on the density of liquid in the syphon. There is a significant difference in hydrostatic pressure caused by the difference in liquid column height. The difference in air pressure is actually in the WRONG direction! Air pressure is higher at the bottom than at the top, which would tend to be pushing liquid UP the tube and not down.

Your juice boxes may well have been filled under conditions of slightly positive pressure inside the machine - to keep the bugs out. But I think that they are, in fact totally filled with liquid so there is no air inside to expand. You would have to examine the sides of the boxes and see just how flat they are and whether they 'sag' when left to themselves. This would cause excess pressure inside due to the fact that the top level would have dropped by a mm or two.

In any case, they are filled in a holder with flat sides and resting on a flat surface. I reckon that, when you pick a box up to open it, the pressure of your hand around it (at the contact points with your hand) may well be higher than AP. Can you be sure that you are not just 'too rough' to guarantee no extra force on the box?

They do, indeed, "sag" when left to themselves, as I've previously referred (on another thread) to their generally "crushed" appearance when I get them. The flow does not begin until a minute or two after I stick the straw in (whether I suck on the straw or not!), so whatever transient pressure my hand applies when I stick the straw in appears not to be the source of the motive power in this system. Also, I hardly believe that the differential ambient atmospheric pressure between the top and the bottom of a four-inch-tall juice box might be significant to the point of generating, by itself, a continuous flow of juice through the straw.

 

[sophiecentaur]

It's not the "ambient atmospheric pressure". As I said, its hydrostatic pressure inside that distorts the box. As the box changes shape, its volume decreases a bit (the sides are no longer flat and it's lower in height than it was. The juice that is eventually forced up the tube represents the decrease in volume inside the gradually distorted box.

 

[BadBrain]

It's not the "ambient atmospheric pressure". As I said, its hydrostatic pressure inside that distorts the box. As the box changes shape, its volume decreases a bit (the sides are no longer flat and it's lower in height than it was. The juice that is eventually forced up the tube represents the decrease in volume inside the gradually distorted box.

What you're referring to is the incompressibility of water. I can definitely see how the pressure of my hand as I insert the straw might force the already compressed liquid out of the box, but this would most likely be expressed in the form of a spurting of the liquid from the box as soon as the straw pierces it, which is not the phenomenon I've observed. Likewise, the differential resistance to hydrostatic pressure between the cardboard walls of the box and the outlet provided for the straw should provide a stream of liquid emitting from the straw sooner than I've observed (though I may be wrong on this point). What I've observed is the stream of flow from the box approximately a minute or two after having pierced it with the straw, without my having drawn from the straw. This seems to me, intuitively, to have more to do with capillary action than with pressure differential, although both are most likely involved in the phenomenon I've observed.

 

[sophiecentaur]

Capillary cation in a drinking straw would only bring the liquid up by a mm or two, max. I'm sure it's a 'sagging' phenomenon.

 

[Ken G]

Getting back to how a siphon works, I'm skeptical that it is capillary action. Instead, I would imagine that sophiecantaur has it right-- the force that pushes the liquid through the tube is the pressure difference caused by the extra weight of the liquid compared to air at any height below the top level of the liquid. The way to argue that siphons are not driven by atmospheric pressure is to argue that they would also work on the Moon, for liquids that are enclosed but under only the pressure of their own weight. The need for internal pressure to do the job is evident from the fact that a siphon only works if the tube is full of liquid. Of course one does need suction to fill the tube initially, and atmospheric pressure helps a lot there, but the continued action of the siphon has nothing to do with atmospheric pressure.

 

[sophiecentaur]

Getting back to how a siphon works, I'm skeptical that it is capillary action. Instead, I would imagine that sophiecantaur has it right-- the force that pushes the liquid through the tube is the pressure difference caused by the extra weight of the liquid compared to air at any height below the top level of the liquid. The way to argue that siphons are not driven by atmospheric pressure is to argue that they would also work on the Moon, for liquids that are enclosed but under only the pressure of their own weight. The need for internal pressure to do the job is evident from the fact that a siphon only works if the tube is full of liquid. Of course one does need suction to fill the tube initially, and atmospheric pressure helps a lot there, but the continued action of the siphon has nothing to do with atmospheric pressure.

A syphon would work perfectly well on the Moon (I think you mean in a vacuum but with some Gravity) You would need to use a liquid with a very low vapour pressure - such as oil or mercury but not water - which would turn to vapour and fill the top of the inverted tube.

Capillary action is a definite no-no for this.

 

[Ken G]

Actually, on further thought, I realize that atmospheric pressure is indeed essential for a standard siphon! The problem is, you need to get the pressure difference to push the fluid up to the top of the curve in the tube, which for a standard siphon generally goes above the fluid's maximum level. To get the fluid to flow "over the hump" is where atmospheric pressure is essential, because the weight of the fluid alone won't push the fluid above it's highest level. So the dictionary is right-- it's atmospheric pressure that is the key.

 

[sophiecentaur]

Yes, you are right, of course (been there before and I should have remembered). You still need AP to support the shorter column.
AP is the key but it's not the difference in AP pressure at top and bottom that governs the rate of flow - the difference in hydrostatic pressure has a massive effect on that.

Is this really really finally finally ... . .I wonder.

 

[russ_watters]

We've had this thread before. The liquid won't flow through the siphon - it won't even stay in the siphon - without air pressure.

 

[BadBrain]

We've had this thread before. The liquid won't flow through the siphon - it won't even stay in the siphon - without air pressure.

I apologize for the schematic nature of this response, but I had a much better worked-out version of this response, which got wiped due to the fact I forgot to copy it to my computer before attempting to post (why is the sign-on for this forum of such brief duration?.)

And, at the time I attempted to post that item, I was not yet aware of your post, russ_watters.

I was beginning to think of "air pressure" more in terms of localized partial pressures of the component gasses within the local system, such that gasoline is volatile to the point that it replaces the air within an enclosed space (such as a car's gas tank) possibly providing a pressure upon the surface of the liquid gasoline greater than that which the ambient air could provide on its own. This would not be enough to force the gasoline "over the hump", which is why the assistance of suction at the exit end of the tube is required. The same could be said for water, which is certainly less volatile than gasoline, but would still make the air which absorbs it denser than dry air as long as the air is not allowed to expand in volume (although this may cause precipitation (such as the rain storms inside some voluminous buildings (e.g, the Vehicle Assembly Building at the Kennedy Space Center))).

What I'm suggesting here is a more complex picture in which localized gaseous pressure (externally regulated by ambient atmospheric pressure), gravitation, and capillary action all play a role, in degrees varying according to the molecular constitution both of the liquid being siphoned and of the gasses lying immediately upon the surface of that liquid.

 

[allitup]

There is a test we could do, if we had the stuff: Do a siphon experiment on the International Space Station. Will it siphon in zero or very low gravity?

 

[BadBrain]

There is a test we could do, if we had the stuff: Do a siphon experiment on the International Space Station. Will it siphon in zero or very low gravity?

That's a fine test, but I've already suggested a much cheaper one: see if it works with pure ethanol (which apparently exhibits no hydrogen bonding/surface tension/capillary action).

I'm going to try and get my hands on some, as well as on some plastic surgical tubing, but I can't promise results before tomorrow night, as today's a holiday (which doesn't block my access to pure EtOH, rather, just to the plastic surgical tubing!).

***

Actually, siphoning obviously can't work in a zero-G environment, and I've never said gravity has no role in siphoning. I'm just trying to discover the primary source of motive power within a siphoning system amongst three variables, anyone of which may be dominant within any system of differing chemical/physical molecular constitution.

 

[sophiecentaur]

There is a test we could do, if we had the stuff: Do a siphon experiment on the International Space Station. Will it siphon in zero or very low gravity?

You could end up with a lot of condensation on the Windows.

 

[elegysix]

I like to think of the hump or highest point in the hose as containing the "activation energy" of the system - like in a chemical reaction. Once that energy barrier has been overcome, the system will progress forward to a minimum energy level, minimizing gravitational potential energy.

That initial energy could come from anywhere, although most commonly from creating low pressure in the tube - however I had a water bed at one point, and in order to drain it I simply pushed on the bed to start the water flowing through the hose and out my window. Air pressure had nothing to do with starting the siphon nor continuing it in this case.

Of course, as with a falling rock, all the work is done by gravity - with the exception of providing the initial energy needed to start the flow - which is the same as raising the rock in order to drop it.

 

[sophiecentaur]

Are your house and garden in a total vacuum then?

 

[russ_watters]

This would not be enough to force the gasoline "over the hump", which is why the assistance of suction at the exit end of the tube is required.

Are you familiar with the concepts of gauge and absolute pressure? Because the problem here is that you are misusing them.

Let me restate my previous post in another way: a siphon will not function in a vacuum. There can be no "suction" in a vacuum to fill the siphon.

You also need to learn about capillary action. It's irrelevant here, but you need to learn about it so you will accept it.

 

[BadBrain]

Are you familiar with the concepts of gauge and absolute pressure? Because the problem here is that you are misusing them.

Let me restate my previous post in another way: a siphon will not function in a vacuum. There can be no "suction" in a vacuum to fill the siphon.

You also need to learn about capillary action. It's irrelevant here, but you need to learn about it so you will accept it.

Thanks. I'll brush up on them. It's been awhile since I've been involved in physics (by which I mean decades), and I'm no doubt rusty.

 

[BadBrain]

russ-watters:

OK, I see it now. The initial suction creates a pressure differential which gives the inside of the tube a negative gauge pressure, . The negative gauge pressure within the tube then continues the suction within the tube (assisted by the ambient pressure exerted upon the surface of the gas in the tank), as a flowing fluid will tend to exert a lower internal pressure than does a static fluid, while gravity gives an assist in preventing ambient atmospheric pressure from blocking the hose at it's outlet in the jerry can.

Did I get it right this time?

(This is actually quite close to my initial criticism of Dr. Hughes' article. I wonder how I was convinced away from it?)

 

[sophiecentaur]

"Suction"
This would need cohesion between the water molecules so the downhill ones could, like a chain, 'pull' on the uphill ones upwards. I have seen a movie which sort of implied that cohesion was, in fact, possible because a U tube more than 10m appeared to make a syphon with water.
Any opinions? Dynamic cohesion is a bit more acceptable, perhaps.?

 

[BadBrain]

I like to think of the hump or highest point in the hose as containing the "activation energy" of the system - like in a chemical reaction. Once that energy barrier has been overcome, the system will progress forward to a minimum energy level, minimizing gravitational potential energy.

That initial energy could come from anywhere, although most commonly from creating low pressure in the tube - however I had a water bed at one point, and in order to drain it I simply pushed on the bed to start the water flowing through the hose and out my window. Air pressure had nothing to do with starting the siphon nor continuing it in this case.

Of course, as with a falling rock, all the work is done by gravity - with the exception of providing the initial energy needed to start the flow - which is the same as raising the rock in order to drop it.

Actually, I had a response to a similar idea posted earlier on this thread in the post which I mentioned I had to summarize due to my having lost the original as I had timed out of the forum prior to hitting the 'post" button.

What I had originally indicated in my lost post was that I have some familiarity with the Water Supply of the City of New York (yes, that's the name of the agency; I didn't just capitalize "Water" and "Supply" gratuitously), which is fed from two aqueducts (soon to be three, but the third will be far underground and use pumps) whose outfalls are at the level of the average ground-floor level in Manhattan after a drop from the average sixth-floor (fifth-storey in UK usage) level in Manhattan. This creates the typical gravity-driven system of liquid surfaces all being at the same level regardless of the volume of the container. with every vertical container being fed FROM THE BOTTOM, and not having to go over any "hump". The same principal is used by small town with water towers on the outskirts (usually with some advertisement for the town painted on the side).

 

[BadBrain]

"Suction"
This would need cohesion between the water molecules so the downhill ones could, like a chain, 'pull' on the uphill ones upwards. I have seen a movie which sort of implied that cohesion was, in fact, possible because a U tube more than 10m appeared to make a syphon with water.
Any opinions? Dynamic cohesion is a bit more acceptable, perhaps.?

I no longer think so, although I'll still perform my EtOH experiment and report the results.

Mentors: will the little bit of water (about 5%) in a bottle of pure grain alcohol throw the results of such an experiment?

My original formulation (which I now fully recall as having been in terms of blast overpressures and underpressures) was that the flow of the liquid through the tube will create an underpressure within the tube relative to the pressure of the static liquid in the donor container (i.e., a negative gauge pressure) perpetuating suction at the opening of the tube within the donor container.

 

[BadBrain]

You can also think in terms of an airfoil, where the convex curve on top causes the air to flow faster over the top of the wing than it does under the bottom, thus lowering its density and decreasing its pressure, in turn creating a certain amount of suction on the upper surface of the wing and allowing gravity to cause the bottom surface to rest on the denser air beneath the lower surface of the wing.

The reverse is, of course, true concerning a spoiler on a racing car, which is an upside-down airplane wing intended to keep the car on the ground to inhibit its tendency to become airborne, the shape of the fuselage also being something of an airfoil.

 

[Drakkith]

Ok, but what does an airfoil have to do with a syphon? Or are you just elaborating on your statement in post 25?

 

[BadBrain]

Ok, but what does an airfoil have to do with a syphon? Or are you just elaborating on your statement in post 25?

The latter.

 

[Ken G]

"Suction"
This would need cohesion between the water molecules so the downhill ones could, like a chain, 'pull' on the uphill ones upwards. I have seen a movie which sort of implied that cohesion was, in fact, possible because a U tube more than 10m appeared to make a syphon with water.
Any opinions? Dynamic cohesion is a bit more acceptable, perhaps.?

The suction is just the difference in pressure between the atmospheric pressure at the top of the liquid, and the absence of atmospheric pressure in the evacuated tube. There's no need for any interparticle forces, an ideal gas would siphon just fine. Now, real liquids do have some contribution from interparticle forces, I'm just saying it's not an essential feature.

 

[sophiecentaur]

Of course we call the difference in pressures Suction. But that explanation doesn't apply to a syphon that operates above the '10m limit' - if indeed it does.

 

[mrspeedybob]

I think a 10m+ siphon would work, just very slowly. The top of the tube would be full of water vapor instead of liquid.

Suppose the source body has a surface elevation of 10 meters, the tube goes up 20 meters and then down 30 meters to the destination body. Water travels up 10 meters from the source as liquid, then vaporizes. It then travels up another 10 meters and down 20 as vapor. It condenses 10 meters above the destination body. The weight of 10m of liquid + 20m of vapor > 10m liquid + 10m vapor so the siphon keeps going.

May be tricky to get started.

 

[BadBrain]

SAFETY WARNING: Isopropyl Alcohol is toxic and must not be swallowed, nor inhaled, nor allowed contact with the eyes. I only got a little in my mouth, and spat it out and rinsed my mouth immediately.

***

I've just performed my experiment:

I used a store-bought 91% solution of isopropyl alcohol, a miniature (2 oz) martini glass as my donor tank, a miniature (2ox) gravy boat as my receptor tank, and a straw from a juice box for my hose (yes, I actually bought one just for this experiment).

Observations: having initiated flow via oral suction, flow continued as long as the mouth of the short end of the straw remained submerged beneath the surface of the liquid within the donor tank.

Measurements Taken: HEY, C'MON! I was working with tiny little glass vessels and a tiny little drinking straw that I could hardly hold in my hands. If want measurements, YOU take them!

Conclusions (keeping in mind my "Possible Sources of Error" discussion below):

1) Isopropyl alcohol is capable of being siphoned, thus suggesting that intermolecular hydrogen bonding is neither a significant nor a necessary force for the operation of a siphon system;

2) Isopropyl alcohol tastes terrible!

Possible Sources of Error:

!) The physical-chemical properties of ethyl alcohol and isopropyl alcohol may differ with regard to intermolecular attraction;

2) The 9% H2O content may have provided sufficient intermolecular hydrogen bonding to maintain the flow (although it wasn't a drip, but a good, strong flow).

***

Now I'll spend the rest of my life looking in vain for a laboratory that will have me!