Lab Reports

   GENERAL RULES: Lab reports should be produced on a laser printer or ink-jet printer; they must be printed on one side of the page and the text must be double-spaced. To the greatest extent possible condense data and numerical results into tables for easy reading. Tables of data or numerical results are single-spaced and should be produced using the wordprocessor's table routine; columns must be labeled and units specified. Graphs and other figures should be made conveniently small and placed in the text as close as possible to their first mention in the report. Make sure that axes are labeled and units are specified. All figures should have a caption. Take care to use correct superscripts and subscripts when writing chemical and algebraic formulas. All but the simplest mathematical equations should be created using the wordprocessor's equation editor. If you place a chemical structure from ChemDraw into the document, make sure that the image is appropriately sized. All pages of the lab report should be numbered except the cover sheet. The page bearing the title and abstract is page number 1. The final page of your report should contain "Literature Cited".
    Place a cover sheet on top of the report. The cover sheet should bear a title for the experiment, The name of the course, your name, the name of your lab partner, and the date that the experiment was carried out in the laboratory. The title of the experiment should also appear at the top of the first page. Do not place your name or your lab partner's name on any numbered page of the lab report.
    Have your lab partner read your complete lab report and fill out the evaluation form. Read the evaluation and decide if you want to make changes to your report. Then turn in the report and the evaluation form.

* * *

REPORT SECTIONS: I strongly suggest the following outline for your lab reports, although for some experiments certain sections may not be needed. When writing the lab report, remember that the important ideas to communicate are the results of your measurements, the analysis of the data, and the conclusions you can draw based on your results. Many of the sections of the lab report, especially data analysis and sample calculations, will help you think about your results. When you include a section, label it!

Experiment Title

Abstract: A brief paragraph on the question your experiment attempts to answer, and the important results (numerical results) and conclusions you have made from the experiment. Other students should be able to read your abstract and decide whether they are interested in the rest of the report. The abstract DOES NOT contain a description of the experiment or background information behind the experiment.

Abstracts that begin with "The purpose of this experiment was ..." or "We did this experiment to find out ..." are usually inappropriate. These sentences belong in the Introduction. A good starting sentence for an abstract is "Our investigation of [ ] revealed that ..." That start leads directly to results and conclusions.

Introduction: A few brief paragraphs on the reasons for doing the experiment. What is the question about Nature that you hope to answer? Why is the question important? Discuss what you already know and what you hope to learn. Any theory or prior experiment that is related to the current question should be discussed here.

If the molecule under study is big or unusual, like azulene, this is the place to insert a chemical structure from ChemDraw.
If you are investigating a particular physical equation, such as the Nernst equation, insert it in the introduction using the equation editor in Word.

Experimental Method: A brief outline of how the measurements are made. In many cases a nice drawing of the apparatus will be helpful. You should not describe in detail how the experiment was performed; instead, reference the source of your procedure.

Raw Data: A presentation of the measurements you made during the experiment. A table or graph will usually help the reader absorb the data. Place several sets of data on one graph for easy comparison. Show all the data you took. If you reject some of the data from further analysis, explain why.

Data Analysis: Describe the method of analyzing the data that you used to determine any physical or chemical constants. Be sure to include a sample calculation so that others can follow your calculation. If you used a computer program, be sure to mention which program you used. If you wrote the program, include a copy in an appendix to your report.

Never present the numerical results of a least-squares fit without an accompanying graph;
the fit function should be graphed as a line and the experimental data should be plotted as individual points.
Never copy the entire output of an Excel regression into a Word document without editing it to
remove the unneeded words and statistics. You are better off formating the results in Word as a Word table, or creating a new Table in Word. All tables must conform to ACS style: no vertical rules!

Error Analysis: Determine the 95% confidence limits of the physical or chemical constant you found during the data analysis. Compare your constants with the literature values and determine the accuracies (percent error). Did the correct answer fall within the confidence limits you computed? If not, why not?

This is the appropriate place to discuss determinate sources of error; that is, errors that could have thrown your value off either too high or too low. Generally, if the accepted value is outside your 95% confidence limits, then your experiment suffers from some determinate source of error, and you should attempt to find out what it is. Usually a few careful control experiments will help you track down the source of error.

   Conclusions: Bear in mind that your conclusions are the most important part of the report. You do an experiment to answer a question about Nature. The answer to the question should be the focus of your conclusion. The preliminary sections of the report simply provide the reader with the background needed to accept your answer. ANSWER THE QUESTION!
    Use the lab report as an opportunity to think about the physical and chemical implications of your results. When you believe your results answer the question put to Nature, you should try to understand the physical and chemical meaning of your answer: Does the structure of the molecule under study explain the result? In other words, why did you get the answer that you got? Mention other questions that remain unanswered, and propose further experiments that could answer them. Could you do the same experiment on a different compound, or a different experiment on the same compound, that would answer other questions? Most bad lab reports are written because students don't think about the meaning of their results.
    A Common Pitfall: The "Quality of the Data" Conclusion. Your conclusion should not say that you got "perfect" data or good data or acceptable data or bad data. The goal of the experiment is to answer a question about Nature, not to obtain "perfect" data. Getting "perfect" data is impossible! Getting "perfect" data or even good data does not make you a scientist. You are a scientist when you ask questions about Nature and design experiments to answer them. Many students misuse their conclusion in an attempt to explain why their data are bad or untrustworthy, so that they can avoid answering the question put to Nature. This will not be permitted!

Random error exists in all experimental data -- multiple measurements and signal averaging will slowly remove random error from your results. Much more insidious is determinate error -- the way to find out if determinate error exists is to run little control experiments throughout the course of the experiment. Ask yourself at every stage of the experiment, "Does this make sense? Can I believe these results?"

Data that are so "bad" that they do not answer the question will not be accepted. If an experiment is incapable of producing data good enough to answer the question, you must find out why. Therefore, you must repeat an experiment until it answers the question put to Nature, or until you understand why it cannot.

Literature Cited: Be sure to cite the sources of your lab procedure, your data analysis, and any chemical information that you mention in the introduction and conclusion. I am particularly looking for citations to related information you got from physical chemistry textbooks, physical chemistry laboratory textbooks, and articles in the Journal of Chemical Education. As a general rule you should follow the citation conventions that you see in the Journal of Chemical Education. Be aware that these conventions will surely be different from those you had to use in earlier writing classes. Consult the ACS Style Guide or ask me if you have questions!

* * *

   MATTERS OF STYLE: Write your report so that your intended audience is other physical chemistry students. This shouldn't be hard, since you will probably read a few old lab reports before you do the experiment, and remember that your lab partner must read your report.
    Sime has some good recommendations on matters of continuity and style, tense, voice, grammar, numbers, and proofreading. Here's a quick summary:
Introduce tables and figures with one or two descriptive sentences above their location in the report.   Use the text below the table or figure to describe its meaning and significance.
Conjunctive Adverbs such as "above all", "rather", "therefore", and "however" will help your reader connect sentences and ideas. Sime has more examples.
Keep your language simple. Try to use short, common words and simple phrases. Avoid fancy words and fuzzy phrases. Sime has many examples of fancy words/phrases and simple alternatives; here are just a few:
                                        FANCY                                            SIMPLE
                                        ascertain                                           find out
                                        endeavor                                          try
                                        modification                                     change
                                        utilize                                               use
                                due to the fact that                                    because
                                    on account of                                       because
                                    in order that                                          so
                                    it is obvious that                                   obviously
Report your work in the Past Tense.
Use the passive voice throughout your description of the experimental procedure.
"10 mls of the NaOH solution were pipetted into the test tube and shaken until a precipitate formed. The preciptate was collected by filtration on ashless filter paper." Notice that we avoid "I did this" and "then we did that".
Watch out for dangling participles. Passive voice descriptions of experimental procedures can become infected with dangling participles. "Using a 10 ml pipette, NaOH solution was added." Switch to a prepositional phrase or, on some occasions when clarity demands, switch to the active voice. "10 ml of NaOH solution were added with a pipette." OR "Using a 10 ml pipette, we added NaOH solution."
Report published results and accepted theories in the Present Tense.
"The voltage of electrochemical cells is predicted by the Nernst equation."
"Ethanol is a more polar solvent than hexane."
Report your conclusions in the Present Tense.
"Our results show that NaOH is a stronger base than NH₃."
"We found that azulene's heat of combustion is higher than that predicted using simple bond energies."
Use the active voice during the introduction and conclusion. Transitive verbs will help you out of the passive voice trap: "Our results indicate ..."; "My experiments show ..." "Our control experiment proved ..." These are stronger, clearer sentences that make for better reading. (I have no objection to personal pronouns in your lab reports; there's nothing wrong with an occasional "I" or "we". CS)
Proofread a paper copy. Try reading the paper out loud.

Summarized from Rodney J. Sime, Physical Chemistry: Methods, Techniques, and Experiments, Saunders College Publishing, 1990. Chapter 8, pages 165-175.

* * *
CRAZY EXCERPTS:Try to avoid writing nonsense such as these examples of logical slips and brilliant bull, which are taken from real p chem lab reports. I've included the comments I wrote in the margin of the report:

Means testing [of chloride percentage] is used to determine if the [unknown] sample is the same as the known ...
Means testing cannot prove that two substances are identical. Two different chemical compounds could have the same chloride percentage.
... it was evident that our temperature readings were higher than expected.
By how much?
... we can attribute this systematic error to improper thermometer calibration.
Any assumption such as this should be tested independently. Check the thermometer in an ice bath and in boiling water.
    Raw Data
weight of hydrochloric acid = 0.734 g
I know you didn't weigh out the hydrochloric acid solution!
moles HCl = 0.20 moles
0.734 g HCl = 0.0201 mole HCl
Heat capacity assumes ideal gas behavior as well as a reversible gas.
How do you shift gears on an ideal gas?
CO₂'s trigonal planar configuration ...
Carbon dioxide is linear.
    Raw Data
trial     mass                    volume                     calculated molecular weight
1          .0912 g                 37.14 ml                        60.34 g/mole
2          .1018                    42.33                             59.09
3          .1008                    45.45                             54.49
The volume of a gas sample will be proportional to its mass. In the calculation of molecular weight, we divide by volume. Therefore the third molecular weight value is low because the volume is too large.
... acetone could have inhibited vapor molecules from displacing water in the burette, thus resulting in a volume for the third trial that was too low.
The volume's too large, not too small! Your explanation is meaningless!
    Raw Data
trial    barometric pressure    temperature
1        745.74 mm Hg            20.6 C
2        745.74                        20.6 C
3        745.74                        20.6 C
... temperature and barometric pressure were closely watched during the experiment. They remained approximately constant through all trials performed.
Why "approximately"? They never changed at all!
Possible causes of error in this experiment were variations in the barometric pressure, the temperature, ...
What variations? Your data denies the existence of changes in temperature and pressure. You are contradicting yourself.
    Error Analysis
The percent error in this experiment was quite large. It was calculated as follows: % error = [(26.69 - 17.74)/26.69] x 100% = 33.53%. However, it is hard to say how accurate this experiment can be. There exists no value with which to compare our experimental value.
You have just contradicted yourself; percent error is a measure of accuracy.
All three trials fell within the 95% confidence limit.
Completely unsurprising! The important question is, does the accepted literature value fall with in the the confidence limits.
The reason for the large variance is because the trial is so far off from the actual molecular weight, due to a high standard deviation.
Statistical nonsense! The standard deviation is high because the value is far from the mean, not vice versa!
... the average of the three trials differed from the accepted value by only 3.56%. However, when you take a closer look, the results from sample 3 should disturb you. It differs by 9.33% from the mean.
Whoa! Why did you switch from comparing against the accepted value to comparing against the mean?
This shows that the results from the other two trials are good enough to save the combined results.
Statistical nonsense! Results cannot "save" other results.
... because the reciprocal of temperature is taken, a small error in the temperature means a huge error in the reciprocal.
Sounds like what you need is a big error in temperature so you'll get a small error in the reciprocal!
In retrospect, we shouldn't have been so zealous to get as many points on the graph. We should have had fewer points ... More actual points would fall on the best straight line in this case.
Great idea! You should probably get just two points, so that they will both fall exactly on the best straight line.
                                                     *                      *                        *