Forensics and the Environment (BJK) Flashcards
1
Q
Why is metals analysis a possibility and why is it useful?
A
- environmental levels and distributions are controlled by the local geology and anthropogenic intervention
- profiles of metals can provide unique fingerprints of environment or activity
2
Q
What are the applications of metals analysis?
A
- assessment of contaminated lands (harmful metals)
- idenfication of oil spills
- provenance studies (animal and crop products) using trace materials
- origins of worked materials e.g. glass, ceramics
3
Q
Describe how glass is formed.
A
- glass is formed when a melt cools too fast for nucleation to occur, hence it has an amorphous structure - a disordered network
- various metal cations are accommodated within the structure randomly
- metals affect the properties of the glass
4
Q
Describe the composition of glass.
A
Two main components:
- the former - commonly SiO2
- the flux - plant ash / natron; the metals lower the melting point and increase viscosity of the glass
Additional components:
- pigments - transition metals impart various different colours
- opacifiers - alter translucence to produce opaque regions
5
Q
Describe the metal contents of glass.
A
- Major and minor: from principal components of former and flux
- five key oxides
- can reveal origins of the raw materials and different generic types of historical glass
- Minor: pigments
- often transition metals
- Trace: low level species present in all materials used in manufacture
- ultimately reflect the local geology of the sources of the raw materials
6
Q
How are metals analysed?
A
- inductively coupled plasma mass spectrometry (ICP-MS) often employed for multi-element analysis
- analyses > 50 trace elements and their isotopes simultaneously
- wide dynamic range = magnitude of the concentrations you can analyse at the same time
7
Q
Describe laser ablation ICP-MS.
A
- good for precious samples (no need to sacrifice)
- high energy laser
- spatial resolution: 100 - 350 μm spot, point or line analysis
- part per billion (ppb) detection level
8
Q
How does laser ablation ICP-MS work?
A
- initial ablation to remove the surface at the spot you’re looking at (remove contaminants from the environment)
- take another recording at the same spot from slightly deeper
9
Q
Describe the accuracy and precision of ICP-MS and laser ablation ICP-MS.
A
ICP-MS:
- good accuracy and precision
Laser ablation ICP-MS:
- very good accuracy but less precision
- largely due to the heterogeneity of the material but also impacted by the way the analysis is performed
- losses are possible when gas flows in to carry the material, also can get differences in flow rate
10
Q
Describe replication.
A
- repeated measurements
- gives a frequency distribution described by the sample mean and the sample standard deviation
- repeat measurements approximate to a normal (Gaussian) distribution
- symmetric about the mean
- s is the spread of the curve
11
Q
Describe normal distribution.
A
- equation involves population mean (μ) and standard deviation (σ)
- to determine these, measure each member of the population
- whatever the values:
- 68% of the population lies within ± 1 σ
- 95% within ± 2 σ
- 99.7% within ± 3 σ
12
Q
Describe replication in analysis.
A
- repeat observations
- the more repeat observations (n = repeats) the closer they represent a normal (Gaussian) distribution
- a Gaussian distribution is described well by 50 points
- we can estime the population from a sample of n repeats provided we have enough coverage
- estimate μ and σ from s and x̄
- selection of n depends on context of observation
- practical considerations (ease of sampling, speed/cost) often limit choice of n
13
Q
Describe outliers.
A
- outliers are results that appear to differ considerably from the population mean
- often there is a tendency to discard these points in order to obtain a ‘better fit’
- statistical test to determine if a value should be excluded or not = Q test
14
Q
Describe the Q test.
A
- the difference between the questionable result and its nearest neighbour is divided by the spread (w) of the entire data set (including the outlier)
- Qexp is compared with rejection values (Qcritical) at the right confidence level
- if Qexp > Qcritical then reject
15
Q
When/how can the Q test be used?
A
- suitable for limited sample sized (< 30)
- caution when dealing with small samples (lower limit of n = 4)
- must view the statistical test only in the light of the chemistry involved
- rejection is based on awareness of an error in measuring a particular result