High Performance Liquid Chromatography as a valuable tool for geoforensic soil analysis

Figures & data

Figure 1. Aerial view of the sampling site.

Figure 2. Overview of Location 1.

Figure 3. Overview of Location 2.

Figure 4. Overview of Location 3.

Figure 5. Overview of Location 4.

Table 1. Summary of HPLC parameters tested.

Method	Gradient			Injection (μl)
	Time	%Acetonitrile	%Water
Gradient 1	0.0	5	95	10
	90.0	95	5
	95.0	95	5
Gradient 2	0.0	30	70	20
	10.0	30	70
	60.0	80	20
	65.0	80	20
	65.1	98	2
	70.0	98	2
Gradient 3	0.0	40	60	20
	10.0	40	60
	15.0	50	50
	31.0	50	50
	50.0	80	20
	55.0	80	20
	55.1	98	2
	60.0	98	2
Gradient 4	0	45	55	20
	5	45	55
	35	75	25
	38	98	2
	40	98	2
Gradient 4a	0	45	55	50
	5	45	55
	35	75	25
	38	98	2
	40	98	2
Gradient 5	0	45	55	50
	5	45	55
	35	75	25
	38	98	2
	45	98	2
Gradient 6	0	45	55	50
	5	45	55
	35	75	25
	38	98	2
	40	98	2
	45	45	55
	50	45	55

Figure 6. Chromatography obtained by Bommarito et al.²⁰ and using the same instrument parameters for samples collected at each location in Brockwell Park.

Figure 7. Summary of the changes in chromatography observed for sample 4A with the instrument parameters detailed in Table 2.

Figure 8. Solution stability for samples from each sample location after 7 days refrigeration.

Figure 9. Plot of sample groups by retention time.

Figure 10. Plot of peak apexes.

Figure 11. Visual comparison of chromatograms.

Figure 12. CDFA plots for retention time, λ_max and peak height.

Figure 13. SEM images of a rounded grain (top), a sub-rounded grain (middle) and a grain displaying complete grain- breakage.

Figure 14. SEM images of an elongated grain (top) and of the observed fracture patterns (bottom).

Figure 15. Mean (n=5) relative proportion of grain types for each location.

Table 2. Classification of samples by visual comparison of chromatograms.

Anonymous category	Anonymised samples present	Sample points(respectively)	Major peaks present
			3–5 min	17–18 min	20–21 min
A	5, 8, 14, 18, 19	1E, 1B, 1D, 1C, 1A	No	No	No
B	1, 2, 4, 9, 10	4E, 4B, 4D, 4A, 4C	Yes	Yes	Yes
C	6, 7, 9, 13, 20	2D, 2A, 2C, 2B, 2E	No	Yes	Yes
D	3, 11, 15, 16, 17	3C, 3E, 3A, 3B, 3D	Yes	No	Yes

Table 3. CDFA results for retention time.

Eigenvalues for retention time
Function	Eigenvalue	% of variance	Cumulative %	Canonical correlation
1	108.985*	86.8	86.8	0.995
2	11.678*	9.3	96.1	0.960
3	4.901	3.9	100.0	0.911
Wilks’ lambda for retention time
Test of function(s)	Wilks’ lambda	Chi-square	df	Sig.
1 through 3	0.000	81.138	48	0.002
2 through 3	0.013	38.834	30	0.129
3	0.169	15.976	14	0.315
Functions at group centroids for retention time^a
Sample group	Function
	1	2	3
Location 1	15.814	0.685	−0.565
Location 2	−6.920	−2.035	−2.806
Location 3	−2.078	−3.316	2.637
Location 4	−6.816	4.665	0.734

Table 4. CDFA results for λ_max.

Eigenvalues for lambda max
Function	Eigenvalue	% of variance	Cumulative %	Canonical correlation
1	6310.592*	67.1	89.5	1.000
2	3090.277*	32.9	100.0	1.000
3	3.894*	0.0	100.0	0.892
Wilks’ lambda for lambda max
Test of function(s)	Wilks’ lambda	Chi-square	df	Sig.
1 through 3	0.000	165.370	48	0.000
2 through 3	0.000	86.619	30	0.000
3	0.204	14.292	14	0.428
Functions at group centroids for lambda max^a
Sample group	Function
	1	2	3
Location 1	79.606	65.475	−0.182
Location 2	61.940	−74.084	.250
Location 3	−71.922	10.485	2.453
Location 4	−69.624	−1.876	−2.520

^a Unstandardised canonical discriminant functions were used in the analysis.

Table 5. Results of CDFA for peak height.

Eigenvalues for peak height
Function	Eigenvalue	% of variance	Cumulative %	Canonical correlation
1	872.831*	89.5	89.5	0.999
2	90.618*	9.3	98.8	0.995
3	11.866*	1.2	100.0	0.960
Wilks’ lambda
Test of Function(s)	Wilks’ lambda	Chi-square	df	Sig.
1 through 3	0.000	124.606	48	0.000
2 through 3	0.001	63.650	30	0.000
3	0.078	22.991	14	0.060
Functions at group centroids^a
Sample group	Function
	1	2	3
Location 1	−38.663	6.988	1.327
Location 2	0.220	−13.437	2.199
Location 3	2.503	−1.237	−5.310
Location 4	35.942	7.686	1.784

^a Unstandardised canonical discriminant functions were used in the analysis.

Table 6. Summary of the relative proportions of the three main grain types for each sample point.

Location	Sample point	% Type 1a	% Type 1b	% Type 1c
1	A	74.3	14.3	11.4
	B	54.2	10.4	35.4
	C	34.0	48.0	18.0
	D	34.0	52.0	14.0
	E	37.8	44.4	17.8
	Range	34.0–74.0	10.4–52.0	11.4–35.4
	Mean	46.8	33.8	19.3
	%RSD	37.3	58.6	48.7
2	A	40.9	20.5	38.6
	B	36.0	42.0	22.0
	C	30.0	56.0	14.0
	D	12.0	80.0	8.0
	E	25.6	65.1	9.3
	Range	12.0–40.9	20.5–80.0	8.0–38.6
	Mean	28.9	52.7	18.4
	%RSD	38.4	43.1	68.4
3	A	15.4	15.4	69.2
	B	36.1	30.6	33.3
	C	16.0	78.0	6.0
	D	32.7	55.1	12.2
	E	41.5	41.5	17.1
	Range	15.4–41.5	15.4–78.0	6.0–69.2
	Mean	28.3	44.1	27.6
	%RSD	42.2	54.2	92.1
4	A	6.3	21.9	71.9
	B	31.0	12.1	56.9
	C	28.0	58.0	14.0
	D	42.9	44.9	12.2
	E	22.5	62.5	15.0
	Range	6.3–42.9	12.1–58.0	12.2–71.9
	Mean	26.1	39.9	34.0
	%RSD	51.2	55.6	83.1

Bommarito CR, Sturdevant AB, Szymanski DW. Analysis of forensic soil samples via high-performance liquid chromatography and ion chromatography. J Forensic Sci. 2007;52:24–30.

 PubMed Web of Science ®Google Scholar