A person with a stopwatch can take a reading every fifteen seconds if they concentrate. A data logger takes one every second, for an hour, without getting bored — and the interesting part of most reactions happens in the first thirty seconds.
📚 What you need to know
A sensor is an input device: it detects a change in its surroundings and converts it into an electrical signal.
A data logger receives those signals, stores them, and displays the results in real time.
Common sensors measure pH, temperature, pressure and conductivity.
Logging gives a high sampling rate, removes human reaction time, and lets a computer calculate averages and gradients precisely.
A database is a structured collection of data that can be searched, sorted and filtered — melting points, enthalpy values, rate constants, spectra and more.
A model is a simplified version of reality, so every model involves approximation and therefore some loss of accuracy.
Simulations let you explore systems that would be unsafe, too slow or impossible to run in a school laboratory.
Sensors and data loggers
A sensor does one job: it turns something physical or chemical into a voltage. The data logger does the rest — it stores that signal as a number, plots it as it arrives, and hands the whole set to a computer at the end.
That last point is easy to overlook. One person cannot watch a thermometer and a pH meter at the same moment; a logger records both against the same clock.
What logging actually buys you
It is tempting to say a data logger is “more accurate”. That is not quite the claim. What it really gives you is far more readings, taken at known, even intervals, with no reaction-time delay between the event and the record.
That matters most where the rate is changing fastest. Plot a gas volume against time and the steep part is over in half a minute; sample by hand every fifteen seconds and you get two points on the interesting bit.
Notice that neither set of readings is wrong. The hand-taken points sit on the curve perfectly — there are just not enough of them to show its shape where it matters.
This is the honest way to justify a data logger in an investigation. Do not write “it is more accurate”. Write that it takes readings every second, so the initial rate can be found from a properly defined curve rather than from two widely spaced points.
The four sensors worth knowing
Sensor
What it measures
Typical use in chemistry
pH probe
how acidic or alkaline a solution is
following a titration curve; testing a buffer
Temperature probe
the temperature of a system or reaction
calorimetry; watching an exothermic reaction peak
Pressure sensor
the pressure of a gas or liquid
gas law experiments; a reaction producing gas
Conductivity sensor
electrical conductivity, which depends on ion concentration
following a rate by the change in ion concentration
How a pH probe actually works
It is worth knowing, because “it measures pH” is not an explanation. A pH probe carries two electrodes. One is a reference electrode, which produces a constant, unchanging potential whatever the solution. The other has a thin glass membrane whose potential depends on the concentration of hydrogen ions outside it.
The meter measures the potential difference between the two and converts it into a pH value, usually to two decimal places. Since the reading is a voltage, it can be fed straight into a data logger — which is how a full titration curve is recorded automatically.
A pH probe must be calibrated against buffer solutions of known pH before use, and rinsed with distilled water between solutions. An uncalibrated probe gives readings that are precise to two decimal places and wrong in all of them — the classic systematic error.
Data you do not have to measure
Not every number in an investigation comes off a sensor. A database is a structured collection of data that can be searched, sorted and filtered quickly, and chemistry has an enormous number of them.
Formulae and charges of polyatomic ions, for building formulae and equations.
Thermodynamic data: enthalpy changes, entropies, Gibbs energy values.
Kinetic and equilibrium data: rate constants and equilibrium constants.
Spectroscopic data: NMR, IR and mass spectra for identifying compounds.
Structures, bond lengths and published synthesis procedures.
Large public examples include PubChem and the NIST WebBook; your data booklet is a small, curated database of the same kind. The important habit is treating a looked-up value as literature data — something to compare your result against, not something to quote instead of measuring.
Models and simulations
A model is a simplified version of reality. Ball-and-stick models represent atoms as spheres and bonds as rods, which is enough to show geometry and bond angles and says nothing at all about electron density. That trade is the point: every model, from a plastic kit to a supercomputer climate run, involves approximation and simplification, and therefore some loss of accuracy.
A simulation uses a model to predict how a system behaves, and lets you change variables and watch the effect. Its usefulness depends entirely on how good the underlying model and its assumptions are.
The obvious school example is the Group 1 metals. Caesium and water is far too vigorous to demonstrate, so the trend is predicted from the metals higher in the group and confirmed by simulation.
WORKED EXAMPLE
A student is following the reaction of magnesium with hydrochloric acid by collecting gas. Explain two advantages of using a pressure sensor and data logger rather than a gas syringe and stopwatch.
Advantage 1 — the sampling rateThe logger records many readings per minute at exactly even intervals, so the steep early part of the curve is properly defined.a reliable initial rate can be foundAdvantage 2 — no reaction timeThe reading and the timing happen together in the instrument, so there is no delay between seeing the volume and stopping the watch.removes human reaction time as a source of errorA third one if you need it: the readings go straight into a spreadsheet, so there are no transcription mistakes.
WORKED EXAMPLE
Explain how a pH probe produces a reading, and state one precaution needed before it is used.
The two electrodesOne is a reference electrode with a constant potential. The other has a thin glass membrane whose potential depends on the hydrogen ion concentration in the solution.The measurementThe meter measures the potential difference between them and converts that voltage into a pH value.the reading is a voltage, converted to pHThe precautioncalibrate the probe with buffer solutions of known pHOtherwise every reading carries the same offset — a systematic error that repeating will never reveal.
WORKED EXAMPLE
A student uses a simulation to predict how vigorously francium would react with water. Comment on how much confidence should be placed in the result.
Why a simulation is used at allThe reaction cannot be carried out safely, and francium is not available in usable quantities, so direct measurement is not an option.What limits the confidenceA simulation runs on a model, and every model simplifies. Its output is only as reliable as the assumptions built into it and the data used to construct it.treat it as a prediction, not a measurementThe prediction is still worth having: it follows a trend confirmed experimentally for the metals above francium in the group.
💡 Exam tip
Define a sensor as an input device that converts a change into an electrical signal.
Justify a data logger by sampling rate and the removal of reaction time, not by a vague claim of accuracy.
Name the quantity a sensor measures and the reaction it would suit; questions usually want both.
Mention calibration whenever a pH probe or colorimeter appears.
Say a model is a simplification, so it has limits — that sentence is worth a mark in evaluation questions.
Use database values as literature comparisons for your own result.
⚠️ Common mix-up
Saying a data logger is “more accurate”. It is more frequent and better timed; a badly calibrated probe is still wrong.
Confusing the sensor with the logger. The sensor detects; the logger stores and displays.
Treating a simulation result as experimental evidence. It is a prediction from a model.
Quoting a database value as your own result instead of comparing your measurement against it.
Forgetting that a pH probe needs calibrating, then reporting values to two decimal places as if that made them right.
Up next: Using Technology to Process Data — a logger can produce several thousand readings in a single run. That is only an advantage if you have a way of turning them into a graph and a gradient.
Want this explained one-to-one?
Book a free session with an experienced IB Chemistry tutor and get your trickiest topics made simple.
