Scientific Writing Skills for MSc Chemistry Students
Most MSc chemistry students are never taught to write. They are taught reactions, mechanisms and instrumentation, and then handed a dissertation deadline. The result is predictable: good laboratory work described so vaguely that an examiner cannot tell what was actually done. Writing is not decoration on top of the science — at MSc level it is part of the science, and it is the part that decides your dissertation marks, your viva, your PhD application and every research email you will ever send.
This guide covers what scientific writing actually means for a chemistry postgraduate, the structure examiners expect, how to report numbers honestly, and the specific errors that lose marks. It is deliberately practical. Nothing here requires a special talent for English — it requires a method.
What you will actually have to write
The MSc "writing load" is wider than the dissertation:
- The dissertation or project report — the big one, usually the largest single piece of assessment in the final year.
- The experimental section — a separate skill from the rest of the report, and the one students get worst.
- A conference or poster abstract — a few hundred words, judged hard.
- A statement of purpose for PhD applications, which is scientific writing wearing a different hat.
- Emails to potential supervisors — the shortest and most consequential writing you will do all year.
The one sentence pattern that fixes most results writing
Weak results writing is usually a sentence missing information rather than a sentence with bad grammar. Use this pattern and most of the problem disappears:
"The yield was good" fails all four. "The isolated yield was 72% after column chromatography (silica, 20% ethyl acetate in hexane)" satisfies them. The reader now knows what was measured, how much, and under what conditions — and could repeat it.
IMRaD — the structure examiners are reading for
Almost every chemistry report follows IMRaD: Introduction, Methods, Results and Discussion. Knowing what belongs in each section stops you writing the same content three times, which is the single most common structural fault in MSc dissertations.
| Section | The question it answers | What must NOT go here |
|---|---|---|
| Introduction | Why does this problem matter, and what is already known? | Your own results |
| Methods / Experimental | Exactly what did you do, in enough detail to repeat? | Interpretation, justification |
| Results | What did you observe? Numbers, spectra, tables. | Claims about mechanism or significance |
| Discussion | What do the observations mean, and what are the limits? | New data appearing for the first time |
| Conclusion | What is now known that was not known before? | Anything you did not show |
The experimental section — write it so a stranger can repeat it
The test is simple: could a competent MSc student in another department reproduce your work using only this section? That means quantities with units, molar amounts, source and grade of reagents, apparatus, temperature, time, work-up, purification method, and the characterisation data with the instrument and solvent stated. Write it in the past tense and impersonally, and write it the same day you do the experiment — not four months later from a half-filled notebook.
Before: "The reaction mixture was stirred for some time and then worked up. The product was purified and characterised."
After: "The mixture was stirred at 25 °C for 6 h under nitrogen. It was quenched with saturated aqueous NH₄Cl (10 mL), extracted with ethyl acetate (3 × 15 mL), dried over anhydrous Na₂SO₄ and concentrated under reduced pressure. Purification by column chromatography (silica gel, 15% ethyl acetate in hexane) gave the product as a white solid. ¹H NMR spectra were recorded in CDCl₃."
The second version is not longer because it is padded. It is longer because it contains the information the first version was missing.
Reporting numbers: mean, standard deviation and significant figures
A single measurement is not a result. If you titrated five times, report the average and the spread — and say how many replicates the average came from. Here is the full working on a real set of burette readings.
Titre values (mL): 24.85, 24.90, 24.95, 24.80, 24.90
Step 1 — mean. Sum = 24.85 + 24.90 + 24.95 + 24.80 + 24.90 = 124.40 mL. Mean = 124.40 ÷ 5 = 24.88 mL.
Step 2 — deviations from the mean.
24.85 − 24.88 = −0.03 | 24.90 − 24.88 = +0.02 |
24.95 − 24.88 = +0.07 | 24.80 − 24.88 = −0.08 |
24.90 − 24.88 = +0.02
Step 3 — square and add.
0.0009 + 0.0004 + 0.0049 + 0.0064 + 0.0004 = 0.0130
Step 4 — divide and take the square root. Using the sample standard deviation (divide by n − 1 = 4): 0.0130 ÷ 4 = 0.00325, and √0.00325 = 0.057 mL.
Step 5 — relative standard deviation. (0.057 ÷ 24.88) × 100 = 0.23%.
How to write it: "The mean titre was 24.88 ± 0.06 mL (n = 5, sample standard deviation, RSD 0.23%)." The uncertainty is quoted to the same decimal place as the mean, and the mean is quoted to the precision the burette actually delivers — two decimals, not five.
Textbooks genuinely differ here, so state which one you used. Dividing by n − 1 gives the sample standard deviation, the correct choice when your five titres are a sample from all possible titres — this is the usual convention in analytical chemistry. Dividing by n gives the population standard deviation; for this data set that would be 0.0130 ÷ 5 = 0.0026 and √0.0026 = 0.051 mL. The two answers are different, so never leave the reader guessing which one you calculated.
References, plagiarism and academic integrity
Cite from the paper you actually read, not from a citation you found inside another paper. Use one referencing style consistently — chemistry departments in India commonly ask for the ACS style, but follow whatever your own department specifies, because that is what the examiner is checking against. Build the reference list as you write; reconstructing it the night before submission is how wrong page numbers and dead links get in.
On plagiarism: rewriting someone else's sentence with a thesaurus is still plagiarism. The honest technique is to read the source, close it, write the idea in your own words from memory, then reopen the source to check you got the science right — and cite it. Indian higher education institutions operate under the UGC's academic-integrity and anti-plagiarism regulations, and each university additionally sets its own similarity policy and penalties. Read your own institution's policy document before you start writing, not after a report is flagged.
A four-week plan for a dissertation you have not started
Week 1 — skeleton and experimental. Write every heading and sub-heading first. Then write the entire experimental section from your notebook. It is the easiest section because you are not arguing anything, and finishing it early kills the worst source of panic.
Week 2 — results. Make every table and every data figure. Write one plain sentence under each describing what it shows, with no interpretation. Do the statistics now, not later.
Week 3 — discussion, then introduction. Write the discussion while the results are fresh. Write the introduction last among the main sections — only now do you know what the report needs to introduce.
Week 4 — abstract, conclusion, references, and one full read-aloud pass. Reading the whole thing aloud catches broken sentences that silent reading skips. Leave at least two days between the last edit and submission.
Mistakes that cost marks
- Overclaiming. "This proves the mechanism is SN1" when your data is consistent with SN1 but does not exclude alternatives. Write "consistent with", "suggests", "supports" — and say what would be needed to prove it.
- Missing units and wrong significant figures. A calculator gives 24.88000; your burette does not. Quote the precision the instrument can actually deliver.
- Interpretation in the results section. Keep observation and explanation apart; it is the fastest way to make a report read as professional.
- Figures and tables with no self-contained caption. A caption must make sense on its own — an examiner reads captions first.
- Copying a method from a paper without adapting it. If you used 50 mL and the paper used 500 mL, write what you did.
- Writing everything in one weekend. Scientific writing is edited, not typed. First drafts are supposed to be bad; the marks come from the second and third pass.
- Passive voice everywhere, on principle. The experimental section is conventionally impersonal, but in the discussion "we propose" is clearer and is accepted in modern chemistry journals. Clarity beats a rule you half-remember.
Where each skill pays off
| Where | What is really being assessed | Skill to prioritise |
|---|---|---|
| MSc dissertation | Can you separate evidence from claim? | IMRaD discipline, honest reporting |
| Viva / project defence | Do you understand your own numbers? | Statistics, error and units |
| Conference abstract | Can you compress a result without distorting it? | The results-sentence pattern |
| PhD statement of purpose | Can you explain your work to a non-specialist chemist? | Structure and plain language |
| Email to a supervisor | Are you specific and easy to reply to? | One clear ask, no padding |
Doing the statistics for your results section. The Mean / Median / Standard Deviation tool computes the mean, standard deviation and spread of a set of replicate readings so you can check the working you did by hand — exactly the calculation shown above.
Open the Mean / Median / Standard Deviation Calculator →Preparing for IIT-JAM, GATE, CSIR-NET or CUET-PG alongside your MSc? ABC Chemistry runs dedicated competitive-exam batches at the coaching centre and online for students anywhere in India — details at abcchemistry.in.