Detection of lithium in scalp hair by time-of-flight secondary ion mass spectrometry with high energy (MeV) primary ions

Karen J. Cloete *a, Boštjan Jenčič b, Žiga Šmit bc, Mitja Kelemen bd, Kwezikazi Mkentane e and Primož Pelicon b
aiThemba Laboratory for Accelerator Based Sciences, National Research Foundation, PO Box 722, Somerset West, 7129, South Africa. Fax: +27 21 843 3525; Tel: +27 82 587 6720E-mail:
bJožef Stefan Institute, Jamova cesta 39, SI-1001 Ljubljana, Slovenia
cFaculty of Mathematics and Physics, University of Ljubljana, Jadranska ulica 19, SI-1000 Ljubljana, Slovenia
dJožef Stefan Postgraduate School, Jamova cesta 39, SI-1001 Ljubljana, Slovenia
eDivision of Dermatology, Groote Schuur Hospital, University of Cape Town, Cape Town, South Africa

Received 3rd July 2017 , Accepted 26th July 2017

First published on 27th July 2017

In this exploratory study, the application of secondary ion mass spectrometry with a heavy primary ion beam with an energy of several MeVs (MeV-SIMS) is presented for the simultaneous detection and spatial distribution mapping of lithium in a single, chemically unprocessed strand of scalp hair across a longitudinal profile.


Lithium detection in biological materials is associated with several challenges.1 The choice of biological medium is particularly important since the stability of the lithium ion signal may be influenced by sample water content.2 Of the biological tissues, hair presents an attractive testing substrate due to its low water content and ease of sampling, handling, and storage. In addition, hair tends to store higher concentrations of trace elements than blood or urine,3 whilst hair also provides a longer diagnostic window depending on the length of the hair shaft.4,5 Besides the choice of biological tissue, selecting an analysis method of high sensitivity, speed, and precision is also critical for lithium detection in complex biological samples.

Hair lithium screening has been traditionally performed using conventional analysis methods.6,7 The most significant issue with such conventional elemental analysis techniques is that sample processing is destructive. Samples from various donors are often pooled and chemically processed, which destroys historical and spatial information on the localized distribution of trace elements preserved within the length of an individual hair strand.4 Chemical extraction methods specific for a proposed set of major elements may also yield low recoveries for trace elements and even introduce contaminants.8

Ion beam techniques have shown great promise for the detection of lithium in intact samples.9 However, a number of experimental factors may require careful optimization or else they significantly influence the outcome of lithium screening. Detection sensitivity may also be greatly affected by sample chemical and physicochemical properties, for example the abundance of heavy elements.

Secondary ion mass spectrometry (SIMS) has been described as a promising ion beam technique for the analysis of lithium.9 SIMS is currently one of the most sensitive surface analysis techniques for chemical mapping, providing micrometer or lower resolution for depth profiling of the first one or two surface monolayers in an intact sample.10 Positive lithium ions required for injection into the mass spectrometer are also easily formed when the sample is irradiated with a primary ion beam, since lithium has a low ionization potential. Furthermore, no destructive or exhaustive sample preparation such as labelling is required for sample preparation.11 To our knowledge, there are however no studies on lithium screening in human hair with SIMS.

In this exploratory study, the application of secondary ion mass spectrometry with MeV primary ions (MeV-SIMS) is presented for the detection and mapping of lithium in a single, chemically unprocessed and longitudinally sectioned strand of scalp hair. Although SIMS is traditionally performed with primary ions in the keV energy range,11,12 we explored the technique with primary ions at MeV energies through a method termed MeV-SIMS.13–15 In MeV-SIMS, the interactions between primary ions and sample molecules are based on electronic energy loss, which creates a softer sputtering process and approximately three orders of magnitude greater yield of non-fragmented secondary molecular ions than those of conventional SIMS. In addition, the application of such an efficient desorption probing beam in the MeV energy range results in negligible surface charging of insulating materials, thus enabling the analysis of an intact surface of hair sample material, in contrast with conventional SIMS, where metallisation is normally used to compensate for the sample charging.

In this study, we demonstrate the high sensitivity of MeV-SIMS for light elements, and its ability to detect lithium in the surface of non-processed biological materials. A schematic illustrating the principle of MeV-SIMS is depicted in Fig. 1.

image file: c7ay01616f-f1.tif
Fig. 1 Schematic overview of the basic principle of time-of-flight MeV secondary ion mass spectrometry.

Results and discussion

Hair samples were collected from the same region of the scalp from participants between the ages of 19 and 22. Collection of the hair samples was completed after ethics clearance was granted by the Research and Ethics Committee at the University of Cape Town, South Africa (HREC/REF: 398/2012) and informed consent obtained from all relevant parties. All applicable international, national, and/or institutional ethical guidelines, such as the 1964 Helsinki declaration and its later amendments, were followed for the care and use of human materials. The collected hair fibres were washed with warm (37 °C) 1% (w/v) sodium dodecyl sulphate to remove hair surface contaminants and rinsed with ultrapure water. The samples were subsequently air dried for a minimum of 4 hours at ambient temperature.

For MeV-SIMS, washed hair fibres were longitudinally sectioned along the hair shaft using the method described in Kempson et al.16 A steel plate (60 × 110 mm) with 5 mm wide grooves of depths ranging between 20 and 80 μm was manufactured and used for longitudinal sectioning of hair fibres. A single hair strand was laid in a groove of fitting depth, secured, and sectioned longitudinally with a stainless steel razor blade. The sectioned hair strands were secured to a silicon wafer with carbon tape in such a manner to expose the incised surface to the primary ion beam. For the morphological investigation of the sample structure, samples were photographed by scanning electron microscopy after the analysis.

MeV-SIMS spectroscopy was executed at the MeV-SIMS setup at the Jožef Stefan Institute (JSI). Samples were irradiated with 35Cl ions, formed in a 2.0 MV tandem accelerator, under vacuum conditions of 2 × 10−7 mbar. Among the charge fractions formed in a stripper channel of the tandem accelerator, a 6+ charge state and an energy of 5.8 MeV were selected.

For mapping mode and molecular imaging of tissues, the beam was focused to a size of 5 × 5 μm2. Focusing was achieved primarily through magnetic triplet lenses, while the final beam size was reached by reducing the collimator and object slit apertures along the beam line. A scanning coil, positioned before the magnetic triplet lenses, raster-scanned the primary ions through the sample surface. Since the sample was rotated at 55° (125°) to the direction of primary ions, the X-span had a larger dimension and the effective scan size was 2.3 × 1.3 mm.

The sample was irradiated for 1 hour with short pulses of 35Cl ions. A pulsed beam was formed by a parallel plate deflector mounted along the beamline. One plate of the beam deflector was constantly biased to 1 kV, while the voltage on the other plate was switching from 0 to 1 kV to create an electric field for guiding primary ions towards the sample in their original trajectory. When the voltages on both plates were matched, the beam was allowed to pass through the beam deflector into the sample chamber. The length of the pulses was approximately 20 ns and their frequency 10 kHz, which totalled 100 μs between each pulse. The frequency of the incoming primary ions bombarding the sample was approximately 3 kHz, indicating that only 30% of the pulses provided a primary ion.

The voltage pulse at the deflector triggered the signal to start the time-of-flight measurements for desorbed secondary ions. A bias of +3 kV in positive ion detection mode applied to the sample holder determined the energy of the desorbed secondary ions, whilst the mass to charge ratio of the secondary ions determined the time between desorption and detection by the microchannel plate detector placed at the end of a 1 m long drift tube. In order to maximize efficiency of the secondary ion detection, an einzel lens is placed at the start of the drift tube. It consists of three electrodes, with the middle one biased to an optimal parameter of 1.5 kV, while the side electrodes were grounded. Both the time-of-flight and the primary beam coordinates were stored for molecular imaging.

The time spectrum from the time-of-flight spectrometer was transformed into positive mass spectra via mass calibration completed using the calibration spectra of the selected reference amino-acids leucine, glycine, and arginine of molecular masses 131 u, 75 u, and 174 u. The feasible mass resolution was estimated at a value of 500 for the m/z value of 400 u, dominantly influenced by the finite length of primary ion pulses. Data processing was performed with the OM Daq program.

The MeV-SIMS analysis has previously been reported as a non-destructive technique15 at low beam fluences. Beam fluence is defined as the number of primary ions hitting the target area unit. The yield of secondary molecules should decrease exponentially as a function of beam fluence (Fig. 2). The parameter which determines the slope of the exponential fall is called the damage cross section and is the analogy of the damage induced on the sample surface by one primary ion. Experiments at the JSI facility with a 5.8 MeV 35Cl6+ primary ion beam bombarding amino acids arginine and leucine, have reported values of damage cross section at approx. 2 nm2. At a fluence of 1012 ions per cm2 – the commonly accepted static limit of the MeV-SIMS which corresponds to 3 hours of measurement under given conditions – approx. 8% of the target surface had undergone chemical alteration due to the ion-induced damage. Since the duration of the hair sample measurement was 1 hour, it is safe to assume that hair samples remained chemically unaltered.

image file: c7ay01616f-f2.tif
Fig. 2 Experimentally measured secondary ion yield of leucine as a function of beam fluence.

Prior to MeV-SIMS analysis, light micrographs were captured for longitudinally sectioned hair fibres. After completion of the MeV-SIMS analysis, the samples were coated with 10 nm of carbon and analysed with a JEOL JSM-5800LV scanning electron microscope (JEOL, Japan). Scanning electron micrographs were captured to assess the efficacy of longitudinal sectioning as well as to support the interpretation of secondary ion maps of lithium. No structural changes in the specimen were noted after the analysis.

The hair samples for MeV-SIMS analysis did not require chemical processing or labelling. Hand-sectioning using a steel plate and stainless steel razor blade delivered longitudinal sections with a level surface containing the internal morphology of the hair fibre for detection and mapping of lithium without embedding. The overlapping cuticle cells and a moderately smooth and fibrous cortical region with minor tearing from sectioning are clearly distinctive in the micrographs of the longitudinally sectioned hair fibre shown in Fig. 3.

image file: c7ay01616f-f3.tif
Fig. 3 Light (top) and scanning electron micrograph (bottom) of a longitudinally sectioned scalp hair fibre shows well-preserved, overlapping cuticle cells and a moderately smooth and fibrous cortical region with minor tearing from sectioning.

The mass spectrum (Fig. 4) from a single, chemically unprocessed and longitudinally sectioned hair strand was obtained with a focused primary ion beam of 5.8 MeV 35Cl6+ ions within a relatively short time frame (within 1 hour). The arrow heads in the figure point to the distinctive peaks of lithium that naturally consists of two stable isotopes; 6Li (6.43%) and 7Li (93.57%). No interfering peaks were observed near 6Li+ or 7Li+. Since a standard-less approach was followed for the analysis, the lithium concentration in hair could not be quantified.

image file: c7ay01616f-f4.tif
Fig. 4 Characteristic spectra of positive fragments in the mass ranges of 0–140 (top) and 0–20 (bottom) acquired from a longitudinally sectioned hair fibre using MeV-SIMS. The arrows indicate the presence of mass peaks for 6Li and 7Li in the spectra.

Besides detection of lithium in hair, MeV-SIMS also allowed mapping of the distribution of lithium in hair across a longitudinal profile that exposed the hair internal morphology. To assess whether the spatial distribution of lithium represented endogenous exposure (localization in the in vivo or endogenous hair regions: medulla and cortex) or exogenous exposure (localization in the exogenous hair regions: cuticle),17 MeV-SIMS imaging was performed on the samples. The distribution map of lithium showed a relatively homogenous distribution of lithium across the entire internal region of the hair shaft (Fig. 5).

image file: c7ay01616f-f5.tif
Fig. 5 MeV-SIMS images showing the homogenous distribution of 6Li (top) and 7Li (bottom) along the longitudinally sectioned hair fibre.

Lithium may enter biological tissues via exposure to water that contains lithium or the consumption of primary food sources such as oats, cacao, seafood, seaweed, goji berries, and egg yolks that contain significant amounts of lithium.18 Although lithium has been considered to have no apparent vital physiological function, nutritional trace element research shows that the biochemical mechanisms of lithium appear to be multifactorial.18,19 At low levels, lithium stimulates proliferation of bone and neural stem cells; acts as an anti-oxidant, anti-inflammatory, and neuro-protective agent; and plays an important role in the transport and uptake of vitamin B12 and folate.18–20 At higher levels, lithium is used as a psychopharmaceutical for the treatment of affective and personality disorders, and to decrease aggressiveness and impulsivity in children with disruptive behaviour, uncontrollable and unstable behaviour in patients with brain injury, and violent behaviour in prisoners.21,22

The reasonably prominent peak of lithium shown in Fig. 4 was only detected in one hair sample originating from a volunteer who may either have been on lithium supplements or psychopharmaceutical drugs. Since the volunteer from whom the hair fibre was sourced was of African ethnicity, it may also be likely that the volunteer previously used alkaline hair straightener or no-lye relaxer emulsions containing lithium hydroxide.23 Use of hair relaxers is particularly popular among women of African ethnicity and is also a deep-rooted cultural practice in many countries.24 “Relaxed” hair is the direct result of an alkaline compound (lithium hydroxide) in hair relaxers that breaks and re-arranges disulphide-bonds of hair keratin molecules.25–27 During treatment, the hair swells, which causes the cuticle scales to open and the relaxer product (lithium hydroxide) to diffuse into the hair from the external environment.

Distinguishing systemic from external lithium exposure remains a challenge as decontamination protocols critical for excluding external contamination may affect the distribution of chemical compounds within hair tissues that differentiate endogenous from exogenous exposure.28 Washing of hair fibres before analysis or during post-relaxer treatment may also not effectively remove lithium ions that may bind to chemical constituents in the hair follicle. Incorporation of lithium into the hair may further be explained by the fact that both lithium isotopes were detected, which may be formed during metabolic processes in the hair follicle. Interestingly, the hair fibre containing lithium was afro-textured, indicating that lithium ion transfer to new hair growth may also be possible. Nevertheless, the exposure model may only be confirmed if past use of hair no-lye relaxers or psychopharmaceutical drugs and nutritional supplements containing lithium are verified.

This is, to our knowledge, the first report to utilize MeV-SIMS for the detection and mapping of lithium and its isotopes in intact and chemically unprocessed hair. We are only aware of one other study that used conventional SIMS to study lithium distribution in biological tissue, but in this case, mouse brain that was subjected to lithium treatment as a neuroprotective agent.29 However, no mention was made of lithium isotope screening. Similarly, this study confirmed that MeV-SIMS is a powerful method for screening and mapping of lithium in biological specimens.


The results of this study show how MeV-SIMS provides a novel, time-efficient, and dual approach for detection and mapping of lithium and its isotopes in biological materials as small as a single hair fibre. Although this approach may present a non-invasive method to determine lithium psychopharmaceutical adherence or lithium nutritional screening, careful consideration is necessary when study participants use hair cosmetics such as relaxers that are known to contain lithium. It has also been cited that lithium levels increase in proportion to dose, reaching a steady state after three months of intake.20 Because of this phenomenon, hair testing for lithium compliance in psychiatric patients should warrant careful consideration. Regardless, SIMS analysis may find application in lithium isotope studies where detection and mapping of lithium in tissues are warranted.

Conflict of interest

The authors declare that there are no financial or other relations that could lead to a conflict of interest.


  1. R. Wuhrer and K. Moran, IOP Conf Series: Materials Science and Engineering, IOP Publishing Ltd, 2016, vol. 109, p. 012019 Search PubMed.
  2. D. A. Lee, J. Am. Chem. Soc., 1961, 83, 1801–1803 CrossRef CAS.
  3. V. Preedy, Hair in Health and Disease, Wageningen Academic Publishers, The Netherlands, 2012 Search PubMed.
  4. I. M. Kempson and E. Lombi, Chem. Soc. Rev., 2011, 40, 3915–3940 RSC.
  5. D. Favretto, S. Vogliardi, M. Tucci, I. Simoncello, R. El Mazloum and R. Snenghi, Forensic Sci. Int., 2016, 265, 193–199 CrossRef CAS PubMed.
  6. J. Schöpfer and G. N. Schrauzer, Biol. Trace Elem. Res., 2011, 144, 418–425 CrossRef PubMed.
  7. I. P. Zaitseva, A. A. Skalny, A. A. Tinkov, E. S. Berezkina, A. R. Grabeklis and A. V. Skalny, Biol. Trace Elem. Res., 2015, 163, 58–66 CrossRef CAS PubMed.
  8. T. F. Abbruzzini, C. A. Silva, D. A. de Andrade and W. J. de Oliveira Carneiro, Rev. Bras. Cienc. Solo, 2014, 38, 166–176 CrossRef CAS.
  9. J. Räisänen, Nucl. Instrum. Methods Phys. Res., Sect. B, 1992, 66, 107–117 CrossRef.
  10. J. C. Vickerman and D. Briggs, TOF-SIMS: Materials Analysis by Mass Spectrometry, Manchester and IM Publications, Chichester, 2nd edn, 2013 Search PubMed.
  11. J. Liu and Z. Ouyang, Anal. Bioanal. Chem., 2013, 405, 5645–5653 CrossRef CAS PubMed.
  12. B. N. Jones, V. Palitsin and R. Webb, Nucl. Instrum. Methods Phys. Res., Sect. B, 2010, 268, 1714–1717 CrossRef CAS.
  13. B. N. Jones, J. Matsuo, Y. Nakata, H. Yamada, J. Watts, S. Hinder, V. Palitsin and R. Webb, Surf. Interface Anal., 2011, 43, 249–252 CrossRef CAS.
  14. L. Jeromel, Z. Siketić, N. O. Potočnik, P. Vavpetič, Z. Rupnik, K. Bučar and P. Pelicon, Nucl. Instrum. Methods Phys. Res., Sect. B, 2014, 332, 22–27 CrossRef CAS.
  15. B. Jenčič, L. Jeromel, N. O. Potočnik, K. Vogel-Mikuš, E. Kovačec, M. Regvar, Z. Siketić, P. Vavpetič, Z. Rupnik, K. Bučar, M. Kelemen, J. Kovač and P. Pelicon, Nucl. Instrum. Methods Phys. Res., Sect. B, 2015, 371, 205–210 CrossRef.
  16. I. M. Kempson, W. M. Skinner and P. K. Kirkbride, J. Forensic Sci., 2002, 47, 889–892 Search PubMed.
  17. I. M. Kempson and W. M. Skinner, Sci. Total Environ., 2005, 338, 213–227 CrossRef CAS PubMed.
  18. T. M. Marshall, J. Am. phys. surg., 2015, 20, 104–109 Search PubMed.
  19. G. N. Schrauzer, J. Am. Coll. Nutr., 2002, 21, 14–21 CrossRef CAS PubMed.
  20. W. Young, Cell Transplant., 2009, 18, 951–975 Search PubMed.
  21. A. A. Nierenberg, S. L. McElroy, E. S. Friedman, T. A. Ketter, R. C. Shelton, T. Deckersbach, M. G. McInnis, C. L. Bowden, M. Tohen, J. H. Kocsis, J. R. Calabrese, G. Kinrys, W. V. Bobo, V. Singh, M. Kamali, D. Kemp, B. Brody, N. A. Reilly-Harrington, L. G. Sylvia, L. W. Shesler, E. E. Bernstein, D. Schoenfeld, D. J. Rabideau, A. C. Leon, S. Faraone and M. E. Thase, J. Clin. Psychiatry, 2016, 77, 90–99 CrossRef PubMed.
  22. C. López-Jaramillo, C. Vargas, A. M. Díaz-Zuluaga, J. D. Palacio, G. Castrillón, C. Bearden and E. Vieta, Bipolar Disord., 2017, 19, 1–49 CrossRef PubMed.
  23. M. F. R. G. Dias, International Journal of Trichology, 2015, 7, 2–15 CrossRef PubMed.
  24. S. A. Aryiku, A. Salam, O. E. Dadzie and N. G. Jablonski, J. Eur. Acad. Dermatol. Venereol., 2015, 29, 1689–1695 CrossRef CAS PubMed.
  25. R. Schueller and P. Romanowski, Cosmet. Toiletries, 1998, 113, 39–44 Search PubMed.
  26. T. C. de Sá Dias, A. R. Baby, T. M. Kaneko and M. V. Robles Velasco, J. Cosmet. Dermatol., 2007, 6, 2–5 CrossRef PubMed.
  27. C. R. Robbins, Chemical and Physical Behavior of Human Hair, Springer, New York, 4th edn, 2013 Search PubMed.
  28. E. Cuypers, B. Flinders, C. M. Boone, I. J. Bosman, K. J. Lusthof, A. C. Van Asten, J. Tytgat and R. M. Heeren, Anal. Chem., 2016, 88, 3091–3107 CrossRef CAS PubMed.
  29. G. Zanni, W. Michno, E. Di Martino, A. Tjärnlund-Wolf, J. Pettersson, C. E. Mason, G. Hellspong, K. Blomgren and J. Hanrieder, Sci. Rep., 2017, 7, 40726 CrossRef CAS PubMed.

This journal is © The Royal Society of Chemistry 2017