Also per the United States Environmental Protection Agency’s Technology Transfer Network Air Toxics Web Site Methanol is listed as a clear colorless liquid with an alcoholic odor, I made ; noted these same observations for my unknown liquid. The key data in identifying my unknown was the boiling point. I found that 5 of 8 non-water liquids listed in the lab manual had densities within 0. 012 g/mL of each other (between 0. 779 g/mL ; 0. 791 g/mL) considering the measured density of my unknown liquid is 0.
7792 g/mL (0. 7827 g/mL graphically) it could very well correspond with any of those 5 liquids, leaving me with a lot of uncertainty.
Whereas the measured boiling point of my unknown liquid is 63. 03 °C which corresponds with only 2 of the possible liquids, methanol (64. 7 °C) ; 2-methylpentane (62. 0 °C). This data was key to quickly ; accurately filtering through a long list of possible liquids. Since 2-methylpentane’s density of 0. 653 g/mL does not correlate well with the measured density of my unknown liquid and methanol’s density (0.
791 g/mL) does correlate with the measured density of my unknown liquid I feel safe in stating that my unknown liquid is methanol. All of the data I collected was useful in identifying my unknown liquid.
The graduated cylinder was the least precise ; accurate of the three instruments, it was difficult to measure a specific amount, as noted by my 2nd trial where I had 31. 9 mL instead of 30 mL. The benefit is its quickness and ease of use. The most accurate instrument was the volumetric pipette, in each trial I was confident that I was adding 10.
00 mL, the downside is it is only used to measure a very specific volume of liquid. The burette had the potential to be the most precise instrument, it has clear volume markings ; the stopper allows a user to dispense very precise amount of liquid.
Unfortunately user experience plays a factor, I felt clumsy using the stopper ; required a lot of patience to properly read the miniscus. If I were to repeat this experiment I would use the volumetric pipette, but once my use of the burette improves I would use that one, for the wider variety of volumes. The uncertainty values listed in the lab manual of 0. 5 mL for graduated cylinder, 0. 01 mL for volumetric pipette ; 0. 04 mL for burette agree with my observations. I do not believe there were any significant sources of error during my experiment to determine the density of my unknown liquid.
As stated in the question the nature of the measuring devices did limit the precision of my measurements. The most obvious limitation was the use of the graduated cylinder which as previously stated was the least precise ; accurate of all instruments, this lead to a third of my collected density values being less precise than the other two-thirds. Another instrument that affected the precision of my density was the scale, the last digit on the scale (the hundredth position) regularly fluctuated between 2-3 numbers, limiting movement around the scale improved the fluctuations, but did not eliminate them.
Since this scale was used for all trials it affected the precision of all densities calculated. In these situations it is impossible to tell the exact effect these limitations had on the calculated densities ; whether the calculated densities would be higher or lower if these limitations were removed. By looking at my results the graphical average density (0. 7827 g/mL) is closer the methanol’s density (0. 791 g/mL) than my numerical average density (0. 7792 g/mL). This does not mean that graphically is more correct than numerically, I just thought it was interesting to note, also a difference of 0. 035 g/mL is not very significant in this experiment considering every instrument used has a greater level of uncertainty. With the graph method once you have your best fit line you can easily estimate the volume that corresponds to a specific mass anywhere along that line (or vice versa), also it’s easier to notice outlier data points when they are graphed. The downside is that unless you use a computer a graph could easily be imprecise ; it takes time to complete. The numerical method is fairly quick and easy, a calculator is all you need.
The graphical approach is great for a larger experiment with many data points, let the computer handle all the work. For a small experiment, like this one, numerical method is perfectly reliable ; much quicker. No data points were excluded. The boiling points were consistent for each trial (63. 0 °C, 62. 9 °C, 63. 2 °C) with a variation of only 0. 3 °C through all the trials I am confident in stating that my methodology led to precise recording of the boiling point. My average boiling point of 63. 03 °C is a bit lower than the literature’s value of 64. 7 °C, which means it is not completely accurate with the literature.
A reason for this could be that the atmospheric pressure in the room was 101. 1 kPa which is slightly lower than the atmospheric pressure for the normal boiling point (101. 3 kPa). A lower pressure leads to a lower boiling point as seen by the formula PV=nrT where P = pressure, V = volume, n = number of moles, r is a constant & T = temperature; since our volume & number of moles stay the same when the pressure goes down the temperature must also go down. This could explain why my average boiling point was not completely accurate with the literature’s value.
I know this formula is used for gas stoichiometry but I believe it is also relevant here. The determination of my boiling point had a greater level of precision than my determination of of the density. As noted earlier through my three boiling point trials there was a variation of 0. 3 °C, or 0. 48% of average boiling point (0. 3 °C / 63. 03 °C * 100 = 0. 48%), whereas my nine density trials had a variation of 0. 018 g/mL, or 2. 3% of numerical average density (0. 018 g/mL / 0. 7792 g/mL * 100 = 2. 3%). Conversely the determination of my density was more accurate then the determination of boiling point.
My calculated average density of 0. 7792 g/mL is 98. 5% (0. 7792 g/mL / 0. 791 g/mL * 100 = 98. 5%) of the noted literature value for methanol, whereas my calculated average boiling point of 63. 03 °C is 97. 4% (63. 03 °C / 64. 7 °C * 100 = 97. 4%) of the noted literature value for methanol. To begin I would eliminate the use of the graduated cylinder, the uncertainty level with this instrument is too high for the readings we are taking, a 10 mL reading has an uncertainty level of 5% which could greatly affect our calculated density.
I would also take more than 9 mass/volume readings, something along the lines of 20 readings would supply a much stronger average density, especially if they were all taken with the more accurate ; precise instruments. Finally if the experiment could be conducted in an area that is regulated to 101. 3 kPa it would lead to a more accurate boiling point reading. All of these conditions put together would lead to a more precise, accurate ; reliable average density. Conclusion: After completion of the experiment I calculated a numerical average density of 0. 792 g/mL and an average boiling point of 63. 03°C for my unknown liquid. Though I do believe that the precision of my calculated data could be improved with better instruments I do not feel that this lack in precision is sufficient to negate the findings I have made. In conclusion I believe that my unknown liquid is in fact methanol which has a known density of . 791 g/mL ; a known boiling point of 64. 7°C. Also, as a secondary conclusion I have found that graduated cylinders are not as precise or accurate for measurements as volumetric pipettes or burettes.