Why did my water contract when exposed to rapid temperature change?

[David89]

Homework Statement

I've been doing an experiment where I measure the water volume in a glass bottle as I manipulate the temperature from room temp (24.8 degrees celsius) to 0 degrees celsius, and then back up to room temp.

I know that water behaves uniquely from 0 to 4 degrees celsius, it contracts instead of expands, but, as the water was increasing in temperature, from 0 degrees to 15 degrees, I placed the bottle in a tub of hot water measuring approx 35 degrees.

Now the water inside the bottle contracted a significant amount before beginning to expand again.

Homework Equations

I have uploaded a pic of my graph, and you can see that at 15 degrees the graph does some weird stuff and then continues on its way

The Attempt at a Solution

I really don't know why the rapid temperature change caused the water to contract initially.

 

[DaveC426913]

It is probably faulty data (it's a single point, easy to put in a wrong digit). Row 48 in your dataset is the culprit.

I am suspicious about the apparent fact that it is a much smaller increment of temperature than all other measurements. All other intervals of temperature measurements are several tenths of a degree; whereas this one is jammed between two other measurements, with only one tenth of the degree on either side, suggesting you took three readings in quick succession. So that data point is anomalous in more than one way. This leads me to believe it has been recorded incorrectly.
But it could also be an artifact of your setup.

Rerun the experiment. I am certain you will see a different result. You will maybe still get the spike, because, if there is a problem with the setup, it probably persists. But it very likely will manifest differently, showing you that it's an artifact, not a bona fide effect.BTW, your graph should probably have height as the Y-axis and temp as the X-axis. Just a preference thing.
That way water height is a function of temperature.
i.e. for any given temperature, there is a unique water height.
(Because, for a function, every point on the X-axis must have no more than one y value).

You cannot say the same as it is.
With temperature as a function of water height, a given water height currently does not have a unique temperature.

 

[David89]

Okie doke,

So I switched the x and y axis, thanks for that suggestion it looks much better that way.
I also performed a curve fit to get a function which reads
0.044x^2 - 0.6174x + 4.9.
This makes sense, when the temp is 0 the reading is 5.2, so I'd say the function is reasonably accurate representation of the data.

Yes I completely agree that row 48 is the culprit. By just looking at the data, this point doesn't make sense. However, this is the point where the bottle was at 15 degrees celsius and increasing at a steady rate. Then the bottle was submerged in 35 degree celsius water and even though the temp went up by a tenth of a degree, which should indicate water expansion, the water contracted down to 4.8 cm very quickly.

More info about the setup:

A cork was inserted to plug the opening of a 414 ml glass bottle. The cork had a pipette inserted through the top to measure the water level as the water in the bottle heated/cooled. A temp sensor was inserted through the top of the cork which displaced the water into the pipette, and my original height was 17 cm with the temp sensor inserted and the temp of the water at 24.8 degrees.

Is it possible that, because the glass is on the outside and makes contact with the water, that the temperature of the glass increases at a faster rate than the water because THAT is what is making contact with the hot water initially? The glass expands, making it appear like the water is contracting, and then when the water reaches the same temperature begins to travel up the pipette again?

I do not believe that it is an error in the setup, I say this because in the experiment it specifically asks why the water level goes down when added to the hot water.

I can include pics in my next post if that helps

 

[Chestermiller]

The first thing that crossed my mind was the thermal expansion of the bottle.

 

[DaveC426913]

Where on the graph is it that you placed the bottle in 35C water? Is that where the anomaly is? Or before?

If the height drop occurred right after submerging the bottle, then yes, expansion would be the culprit.
(I had written out a long response to that effect, then erased it all to first concentrate on the data itself.)

 

[David89]

Yes that is where the anomaly is.

The bottle was taken from room temp and placed in a container filled with hot water when the temperature inside the bottle reached 15 degrees celsius.

How would I include the temperature of the surrounding area as a factor in my graph? I can see what youre saying, to an outside observer it looks random and like there is a mistake there. *The more I think about it the more the expansion of the glass must be the culprit. I didn't realize that it would take some time for the water to heat up and that the glass would be subject to the increased temperature first.

 

[Chestermiller]

Yes. This is correct. One should at least approximately be able to estimate the reduction in apparent water level knowing the coefficient of thermal expansion of the glass.

 

[DaveC426913]

I hadn't realized at first that the glass container was open at the top and that the measurements were essentially above the water level of the bath.
I was picturing a closed vial, under water, which would cause problems with expansion.