# Journal of Forensic Investigation

**Rapid Communication**

**Forensic Application of Atomic Force Microscopy for Age Determination of Bloodstains**

**Threes Smijs**^{1*}, Amu Hosseinzoi^{1} and Federica Galli^{2}

^{1*}, Amu Hosseinzoi

^{1}and Federica Galli

^{2}

^{1}Netherlands Forensic Institute, The Hague, Netherlands^{2}Leiden Institute of Physics, Leiden University, Leiden, Netherlands

**Threes Smijs, Netherlands Forensic Institute, P.O. 24044, 2490 AA, The Hague, Netherlands, Tel: +31 (0)70 888 6516; E-mail: t.smijs@nfi.minvenj.nl**

^{*}Address for Correspondence:**Citation:**Smijs T, Hosseinzoi A, Galli F. Forensic Application of Atomic Force Microscopy for Age Determination of Bloodstains. J Forensic Investigation.2017; 5(1): 6.

## Abstract

## Keywords:

## Abbreviations

## Rapid Communication

**Diagram 1:**Schematic representation of the human bloodstain conditions before and during AFM measurements including an AFM height image for the selection of a RBC for further force measurements.

**Figure 1:**A representative example of a (5-day-old) RBC FV map (A), FD curve (B) and corresponding elasticity map (C) obtained from measurements on a randomly chosen RBC from the peripheral zone of a 5-day-old bloodstain (3 μL, 6.5 - 7 mm in diameter). Mean YM values (± SD) thus determined for 6 RBCs in total are presented in D. The broken arrow in the FD curve (B) refers to the approach curve while the solid arrow in this image indicates the retract curve.The arrows in the image in C indicate areas with artifacts. The bloodstain was deposited on a glass surface in a Petri-dish, dried (23.9 ± 0.5 °C and 35 ± 7% relative humidity, n = 7) and measured after 5 days (27 °C, 36% relative humidity).

**AFM specifications:**Sapphire calibrated silicon tip (Olympus), spring constant 38.75 N/m and sensitivity 18.77 nm/V; Frequency: 300 kHz; Maximum load: 548 to 874 nN; speed: 5 μm/s; Indentation depths: 12 to 16 nm. Statistical data analysis was performed using One-Way ANOVA and Gabriel and Hochberg post-hoc tests (IBM SPSS statistics 20) with a critical level of significance of p = 0.05 and based on YM values calculated [Hertz model; Poisson ratio: 0.5; Tip radii: 10 nm (RBC 1 - 3) and 83 nm (RBC 4 - 6)] for 40564 to 42856 FD curves per cell. *p < 0.001; Size effect η

^{2}= 0.065; Grand mean RBC YM: 1.6 ± 0.4 GPa.

^{2}) effect size appeared to be 0.065. This means that the spreading between the data was for only 6.5% caused by the RBC itself. As illustrated by the representative elasticity map (Figure 1C) the elasticity appeared homogenous over the cell with a grand mean YM for the 6 randomly selected, peripheral bloodstain RBCs of 1.6 ± 0.4 GPa. For comparison it may be noticed that force spectroscopy measurements on well-preserved RBCs from a 5000-year-old mummy tissue resulted in a YM of 2.0 ± 1.0 GPa and for equally processed recent RBCs 2.5 ± 1.2 GPa (sample sizes: 213 - 263). The study used similar AFM conditions (spring constant of 40 Nm

^{-1}, resonance frequencies of 300 kHz, maximum load 500 nN and 10 nm silicon tip radii) as the here presented investigation [9]. As it is not mentioned how many cells were measured with the same cantilever their higher YM values (one would expect lower stiffness for fresh RBCs when compared to 5-day aged cells) could be caused by blunting of the tip. If this is the case the actual, increased radius would render lower values in the applied Hertz model (see Equation 1). As illustrated in Figure 2 increasing the maximum load aggravated this blunting problem. Changes in silicon tip radii were in this study investigated as a function of the number of RBC indentations and sapphire calibrations. An average load of 2812 nN (spring constant 55.1 N/m and sensitivity 14.58 nm/V) resulted in a 50 times increase in radius after 24 RBC indentations and 4 sapphire calibrations (compare Figures 2A with 2D). However, tip radii as large as ~ 600 nm where not used because the resolution of the topography images and the FV data would then be severely reduced. Consequently, potentially interesting nanoscale variations of the YM could thus be missed. For 2 to 6 RBCs indentations the tip radii increased to respectively 27 and 169 nm (Figures 2B and 2C). It may be noticed that observed bluntness is probably mostly caused by the intensive RBC indentations rather than by the few sapphire indentations.

**Figure 2:**Representative SEM images showing changes in tip radius (silicon, Micromash) after a variety of sapphire calibrations (Spring constant: 55.1 N/m;Sensitivity: 14.58 nm/V) and RBC (between 5 and 7 days old) indentations (Maximum load: 2410 - 3213 nN; 256 x 256 pixels).

**A:** Calibrated only once on sapphire; **B:** 1 Sapphire calibration and 2 RBC indentations in total; **C:** 2 Sapphire calibrations and 6 RBC indentations in total; **D:** 4 Sapphire calibrations and 24 RBC indentations in total.

^{2}effect size of 0.81 the effect of this factor is large [10]. Interestingly, η

^{2}first increased from 0.195 to 0.727 for respectively day 5 vs 6 and day 6 vs 7 day but decreased again when YM results obtained for day 7 were compared with those obtained for day 8 (η

^{2}= 0.597). This observation could suggest the start of saturation. It may also be noticed that the mutual spreading between cells within the day increases with increasing bloodstain age indicating a decreased precision of the measurement. An example of the changes that occur in the elasticity maps of a RBC between 5 and 8 days is given in Figure 3B for RBC 3. Similar changes were obtained for the other cells in this stain. It may be noticed that the homogeneous distribution of YM values over the cell disappeared upon ageing of the cells over 5 days while the mean RBC elasticity remained mutual comparable. The periodical bloodstain studies were performed 4 times in total for varying ageing periods (3 to 8 days) under similar controlled conditions, the resulting FV maps processed (Hertz model) into elasticity maps using SEM-supported tip radius corrections, the data pooled and results summarized in Figure 4. These pooled results also showed a significant effect for the bloodstain age in days (p < 0.001) with still a large overall size effect (η

^{2}= 0.611). The size of the effect decreases as the bloodstain becomes older than 5 days. For comparison of 5-to 6-day-old bloodstains η

^{2}is 0.591 while for 6- to-7-day and 7- to 8-day bloodstains η

^{2}was respectively 0.408 and 0.328. Curve fitting of the YM data, illustrated in Figure 4 with the dotted line, resulted in a second-order polynomial relationship between RBC age and elasticity. As it has been reported that RBCs in a bloodstain collapse between 3 and 4 days resulting in a drop and subsequently an increase in YM a non-linear relationship like the one we noticed could be indeed expected when 3- and 4-day bloodstains are included in the measurements [8].

**Figure 3:**Representative changes in YM (± SD) of RBCs selected from the peripheral zone of a bloodstain between 4 and 8 days old (

**(A)**the bloodstain test conditions were the same a described for Figure 2). Representative spatial changes in RBC elasticity resulting from aging of the cell between 5 and 8 days is illustrated in

**(B)**with the elasticity map of RBC 3.

**AFM specifications:** Sapphire calibrated silicon tips (Micromash) with spring constants and sensitivities of 45.5 N/m and 16.64 nm/V, 55.8 N/m and 15.30 nm/V, 55.1 and 14.58 nm/V for respectively day 4, day 5, 6 and day 7, 8; Frequency: 300 kHz; Maximum load: 3018 to 3414 nN; Speed: 5 μm/s; Indentation depths: 20 to 45 nm. Statistical data analysis was performed with One-Way Repeated Measures Anova (IBM SPSS statistics 20) with a critical level of significance of p = 0.05 and based on YM values calculated with the Hertz model (Poisson ratio: 0.5; Corrected radii: 20 nm for day 4 data, 70 nm for day 5 data, 133 nm for day 6 data, 75 nm for day 7 data, 130 nm for day 8 data) for 43046 to 44272 FD curves per cell. Given are the mean YM values/cell (± SD). Size effect η^{2} for the factor day is 0.810. Calculated size effects for day 5 vs 6 , 6 vs 7 and 7 vs 8 were respectively 0.195, 0.727 and 0.597. Grand mean RBC YM: 0.8 ± 0.3 GPa (4 days), 1.3 ± 0.4 GPa (5 days), 2.7 ± 0.3 GPa (6 days), 4.7 ± 0.8 GPa (7 days), 5.5 ± 1 GPa (8 days).

**Figure 4:**Average changes in the YM (± SD) of RBCs obtained from different bloodstains ageing between 3 and 8 days. The dotted line illustrates the polynomial relationship between the data based on curve fitting. Presented mean YM values (Hertz model, Poisson ratio: 0.5) were based on force measurements of 3 RBCs per stain. Prior to and in between the measurements the stains were treated the same as described for Figure 2 and all RBCs were selected from the peripheral zone of the stain.

**AFM specifications:** Sapphire calibrated silicon tips (Olympus andMicromash) with spring constants and sensitivities of respectively 25.16 - 67.48 N/m and 14.74 - 24.34 nm/V; Frequency: 300 kHz; Maximum load: 548 -3905 nN; Speed: 5 μm/s; Indentation depths: 8 to 63 nm.

**Age specification:** 3 days: 1 bloodstain, 139643 FD-curves: 0.65 ± 0.08 GPa; 4 days: 2 bloodstains, 226961 FD-curves: 0.8 ± 0.1 GPa; 5 days: 4 bloodstains,477888 FD-curves: 1.7 ± 0.9 GPa; 6 days: 3 bloodstains, 319896 FD-curves: 2.3 ± 0.6 Gpa; 7 days: 4 bloodstains, 338245 FD-curves: 4.5 ± 0.6 GPa; 8 days: 4bloodstains, 338245 FD-curves: 6.0 ± 1.8 GPa. Statistical data analysis was performed using One-Way ANOVA (IBM SPSS statistics 20) with a critical level of significance of *p* = 0.05.

## References

- Edelman G, van Leeuwen TG, Aalders MC (2012) Hyperspectral imaging for the age estimation of blood stains at the crime scene. Forensic Sci Int 223: 72-77.
- Bremmer RH, de Bruin KG, van Leeuwen TG, Aalders MC (2012) Forensic quest for age determination of bloodstains. Forensic Sci Int 216: 1-11.
- Butt HJ, Cappella B, Kappl M (2005) Force measurements with the atomic force microscope: technique, interpretation and applications. Surf Sci Rep 59: 1-152.
- Stegemann B, Backhaus H, Kloss H, Santner E (2007) Spherical AFM probes for adhesion force measurements on metal single crystals. In: Méndez-Vilas A, Díaz J (Eds), Modern research and educational topics in microscopy. World Scientific Publishing, Spain, pp. 820-827.
- Hertz H (1882) Across the touch solid elastic body. J Pure Appl Math 92: 156-171.
- Heinz WF, Hoh JH (1999) Spatially resolved force spectroscopy of biological surfaces using the atomic force microscope. Trends Biotechnol 17: 143-150.
- Smijs T, Galli F, van Asten A (2016) Forensic potential of atomic force microscopy. Forensic Chem 2: 93-104.
- Wu Y, Hu Y, Cai J, Ma S, Wang X, et al. (2009) Time-dependent surface adhesive force and morphology of RBC measured by AFM. Micron 40: 359-364.
- Janko M, Stark RW, Zink A (2012) Preservation of 5300 year old red blood cells in the Iceman. J R Soc Interface 9: 2581-2590.
- Bakeman R (2005) Recommended effect size statistics for repeated measures designs. Behav Res Methods 37: 379-384.
- Yamanaka K, Noguchi A, Tsuji T, Koike T, Goto T (1999) Quantitative material characterization by ultrasonic AFM. Surf Interface Anal 27: 600-606.