Editing Analytical chemistry (section)

== Signals and noise ==
One of the most important components of analytical chemistry is maximizing the desired signal while minimizing the associated [[Noise (electronics)|noise]].<ref name="isbn0-495-01201-7">{{cite book |author1=Crouch, Stanley |author2=Skoog, Douglas A. |title=Principles of instrumental analysis |publisher=Thomson Brooks/Cole |location=Australia |year=2007 |isbn=978-0-495-01201-6 }}{{page needed|date=January 2014}}</ref> The analytical figure of merit is known as the [[signal-to-noise ratio]] (S/N or SNR).

Noise can arise from environmental factors as well as from fundamental physical processes.

=== Thermal noise ===
{{Main|Johnson–Nyquist noise}}
Thermal noise results from the motion of charge carriers (usually electrons) in an electrical circuit generated by their thermal motion. Thermal noise is [[white noise]] meaning that the power [[spectral density]] is constant throughout the [[frequency spectrum]].

The [[root mean square]] value of the thermal noise in a resistor is given by<ref name="isbn0-495-01201-7"/>
: <math>v_{\rm RMS} = \sqrt { 4 k_{\rm B} T R \Delta f },</math>
where ''k''<sub>B</sub> is the [[Boltzmann constant]], ''T'' is the [[temperature]], ''R'' is the resistance, and <math>\Delta f</math> is the [[Bandwidth (signal processing)|bandwidth]] of the frequency <math> f</math>.

=== Shot noise ===
{{Main|Shot noise}}
Shot noise is a type of [[electronic noise]] that occurs when the finite number of particles (such as [[electron]]s in an electronic circuit or [[photon]]s in an optical device) is small enough to give rise to statistical fluctuations in a signal.

Shot noise is a [[Poisson process]], and the charge carriers that make up the current follow a [[Poisson distribution]].  The root mean square current fluctuation is given by<ref name="isbn0-495-01201-7" />
: <math>i_{\rm RMS} = \sqrt{2 e I\Delta f}</math>
where ''e'' is the [[elementary charge]] and ''I'' is the average current. Shot noise is white noise.

=== Flicker noise ===
{{Main|flicker noise}}
Flicker noise is electronic noise with a 1/''ƒ'' frequency spectrum; as ''f'' increases, the noise decreases. Flicker noise arises from a variety of sources, such as impurities in a conductive channel, generation, and [[Carrier generation and recombination|recombination]] noise in a [[transistor]] due to base current, and so on. This noise can be avoided by [[modulation]] of the signal at a higher frequency, for example, through the use of a [[lock-in amplifier]].

=== Environmental noise ===
[[File:Analyse thermo gravimetrique bruit.png|thumb|upright=1.2|Noise in a [[thermogravimetric analysis]]; lower noise in the middle of the plot results from less human activity (and environmental noise) at night]]
[[Environmental noise]] arises from the surroundings of the analytical instrument. Sources of electromagnetic noise are [[power lines]], radio and television stations, [[wireless device]]s, [[compact fluorescent lamp]]s<ref>{{cite web| title=Health Concerns associated with Energy Efficient Lighting and their Electromagnetic Emissions | publisher=Trent University, Peterborough, ON, Canada
 | access-date=2011-11-12 | url=http://www.emrpolicy.org/science/forum/08_havas_cfl_scenihr.pdf |archive-url=https://ghostarchive.org/archive/20221009/http://www.emrpolicy.org/science/forum/08_havas_cfl_scenihr.pdf |archive-date=2022-10-09 |url-status=live}}</ref> and [[electric motor]]s. Many of these noise sources are narrow bandwidth and, therefore, can be avoided. Temperature and [[vibration isolation]] may be required for some instruments.

=== Noise reduction ===
Noise reduction can be accomplished either in [[computer hardware]] or [[software]]. Examples of hardware noise reduction are the use of [[shielded cable]], [[analog filter]]ing, and signal modulation. Examples of software noise reduction are [[digital filter]]ing, [[ensemble average]], boxcar average, and [[correlation]] methods.<ref name="isbn0-495-01201-7" />