« Nate Vetter - Analytical Problem - Nanoparticles Accumulate in the Food Chain | Main | Allison Blonski-Analytical Problem-Taurine in Energy Drinks »

Heidi Nelson - Analytical Problem: Exposure to Zinc Oxide and Titanium Dioxide Nanoparticles in Sunscreen

Nanotechnology is a rapidly growing field with great potential for many applications. One of the most common commercial uses of nanoparticles is in sunscreen, which often includes zinc oxide (ZnO) and titanium dioxide (TiO2) nanoparticles instead of using them in bulk form. As nanoparticles, these materials are transparent to visible light instead of leaving a white paste on the skin when applied, and they may also be more efficient at scattering UV light.

However, the impact of nanoparticles on human health and the environment is not fully understood. In general, the high surface-area-to-volume ratio of nanoparticles gives them different properties from bulk materials, and their size can enable them to permeate cells. Some forms of TiO2 are photocatalytic, generating free radicals when exposed to UV light and potentially damaging nearby molecules or cells.

To determine whether nanoparticles are safe for commercial applications like sunscreen, it is important to understand what happens when humans are exposed to them. Previous studies have found a small concentration of Zn ions, but no nanoparticles, in blood after exposure to sunscreen containing ZnO. Also, TiO2 nanoparticles have been shown to aggregate in the top layers of skin (stratum corneum) with little penetration deeper into the epidermis.

My hypothesis is that nanoparticles will not be detected in blood, and that both ZnO and TiO2 nanoparticles will only be present in the top layers of skin after sunscreen is applied. I would expect the amount and depth of penetration to depend on the frequency and duration of sunscreen application, nanoparticle concentration, and nanoparticle size. The analytes in this problem are ZnO and TiO2 nanoparticles of varying sizes, and the matrices under consideration are human skin and blood.

1. Biello, D. Do Nanoparticles and Sunscreen Mix? Scientific American, August 20, 2007.
2. Wolf, L. K. Scrutinizing Sunscreens. Chem. Eng. News 2011, 89, 44-46.

Blog 2: UV-vis Absorption Spectrometry

Metal oxide nanoparticles absorb light in the UV range, but they also scatter light. This property makes their UV-visible absorption spectra very complicated. The maximum wavelengths of absorption depend not only on the material, but also on the particle size and shape (3). Metal oxide nanoparticles are also difficult to quantify using UV-vis absorbance spectrometry, since their molar absorptivity also varies significantly with particle size and shape (4). Consequently, there are no convenient reference values for the maximum absorption wavelengths and molar absorptivities, although some size-dependent data and sample values were found.

3a) Bulk ZnO absorbs at 365 nm, and smaller nanoparticles in ethanol absorb at shorter wavelengths (about 340 nm for 5-nm ZnO, and 310 nm for 2.5-nm ZnO) (3). TiO2 nanoparticles of the same size in water absorb in a similar range, 340 to 370 nm (5).

3b) The value of ε for ZnO nanoparticles ranges from 35 m3/kmol mm (350 M-1cm-1) for 2-nm particles to 55 m3/kmol mm (550 M-1cm-1) for 10-nm particles (3). I could not find a value for comparable TiO2 nanoparticles.

3. Segets, D.; Gradl, J.; Taylor, R. K.; Vassilev, V.; Peukert, W. Analysis of Optical Absorbance Spectra for the Determination of ZnO Nanoparticle Size Distribution, Solubility, and Surface Energy. ACS Nano 2009, 3, 1703-1710.
4. Contado, C.; Pagnoni, A. TiO2 in Commercial Sunscreen Lotion: Flow Field-Flow Fractionation and ICP-AES Together for Size Analysis. Anal. Chem. 2008, 80, 7594-7608.
5. Kormann, C.; Bahnemann, D. W.; Hoffmann, M. R. Preparation and Characterization of Quantum-Size Titanium Dioxide. J. Phys. Chem. 1988, 92, 5196-5201.

Blog 3: Similar Analytical Problems

Nanoparticles Accumulate in the Food Chain (Nate Vetter): The hypothesis is that silver nanoparticles in wastewater end up being transferred through the food chain. The analyte is silver nanoparticles and the matrix is wastewater, insects, and animals (not sure of specifics). This problem and my analytical problem both involve detecting nanoparticles in a biological matrix, so we will likely use many of the same techniques. Both types of nanoparticles display size-dependent UV-vis properties. However, the nanoparticle material is different and the particle size may also be different. The behavior of noble metal and metal oxide nanoparticles is likely to be similar but not identical, so some different analytical techniques may be used.

Titanium Dioxide in Masterbatch (Revy Saerang): The hypothesis is that TiO2 particles in Masterbatch affect the uniformity of mixing and pigment distribution. The analyte is TiO2 nanoparticles and the matrix is a mixture of polymer resin, additives, and other pigments. Both of our analytical problems involve TiO2 nanoparticles used commercially, so it is very likely that the same techniques would be used for both problems. However, the context and matrices are different. While Revy is considering how TiO2 nanoparticles affect the properties of a mixture used in industry, I'm looking at whether TiO2 nanoparticles are absorbed through human skin. Detecting TiO2 nanoparticles in Masterbatch may require different techniques from detecting them in blood and skin, especially since skin is a solid.

Blog 6: Chemical structure and Standards

ZnO has the wurtzite crystal structure in its most common form. TiO2 commonly exists in both rutile and anatase structures, although rutile is more stable and consequently more common. Most of the references I have previously cited used wurtzite ZnO or rutile TiO2 nanoparticles, although some extracted nanoparticles from commercial sunscreens and didn't specify the crystal structure.

ZnO (wurtzite)
TiO2 (rutile)

Images from Wikipedia: http://en.wikipedia.org/wiki/File:Rutile-unit-cell-3D-balls.png and http://en.wikipedia.org/wiki/File:Wurtzite_polyhedra.png

ZnO and TiO2 nanoparticles are commercially available in the 10-30 nm size range used in sunscreen:

Nanostructured & Amorphous Materials, Inc.
spherical ZnO nanoparticles (99.5%, average diameter 20 nm) - stock # 5810HT, $70/100 g
needle-like TiO2 nanoparticles (rutile, >98%, 10x40 nm) - stock # 5480MR, $80/100 g

SkySpring Nanomaterials, Inc.
spherical ZnO nanoparticles (99.8%, diameter 10-30 nm), product # 8410DL, $64/100 g
spherical TiO2 nanoparticles (rutile, 99.5%, diameter 10-30 nm), product # 7920DL, $64/100 g


Blog 13? - 1 pt.

Blogs 9, 10, and 11. Good answers.

Blog 11: Capillary electrophoresis techniques

Capillary zone electrophoresis is the only one of these techniques that would work for the analysis of nanoparticles. Although the nanoparticles I am considering are not typically charged, they are not really hydrophobic either, and they are probably too large to partition into micelles, so micellar electrokinetic chromatography would not be useful. Capillary isoelectric focusing is irrelevant to this analyte because pH does not affect the charge and electrophoretic mobility of the nanoparticles (except that they would likely aggregate and crash out of solution at a certain pH, which is not useful for this analysis). Although separating the nanoparticles by size would be useful, capillary gel electrophoresis relies on long, coiled molecules that stretch out to pass through gel pores, and nanoparticles could not change shape to pass through pores and would just get stuck.

In order to apply capillary zone electrophoresis to the nanoparticles, they would need to be coated or functionalized with charged surface groups. Many nanoparticles used in sunscreen are already coated with materials such as silica, aluminum hydroxide, and/or dimethicone polymers, which have a silicon-oxygen backbone with methyl groups (6,7). These coatings would give them a negative charge at neutral pH due to the deprotonation of –OH groups.

However, the use of capillary electrophoresis would require the nanoparticles to be extracted from the solid skin matrix without damaging their coatings, or for them to be re-coated after extraction. The best sample preparation procedure I have found for separating nanoparticles from skin is microwave dissolution (Blog 8), which breaks down organic/biological components and makes it unnecessary to erform any further purification before ICP-MS or ICP-AES. Consequently, chromatography and electrophoresis techniques are not relevant to my analytical problem.

If I were considering nanoparticles from sunscreen in a liquid matrix, I would need to make sure they were charged, either with the coatings mentioned above or by functionalizing them in solution with another charged molecule. Since the nanoparticle coating structure is similar to the glass walls of the capillary, the capillary walls would probably need to be coated with a different substance to prevent their silanol groups from reacting with the particles. I could use a fluorescence detector (similar to Blog 5, but not attached to a microscope), which would enable the detection of individual particles.

Blog 10: Problems using similar techniques

1. The single best technique for my analytical problem is ICP-MS. This technique is sensitive enough to detect small amounts of my analyte (as Zn and Ti). An internal standard can be used for quantification. Microwave dissolution (Blog 8) and plasma atomization will break down most of the organic matrix components, and the ones that remain are unlikely to interfere with the MS signals from Zn and Ti.

2. Other analytical problems using ICP-MS are:
Revy Saerang - Titanium Dioxide in Masterbatch
Sara Baldvins - Geochemical Mobilization of Arsenic to Ground Water
Joe Zibley - I-131 in Japanese Milk Supply
All of these analytical problems involve detecting inorganic analytes in organic or biological matrices.

Blog 9: Chromatographic Techniques

1. Size-exclusion chromatography is the only type of chromatography that would be suitable for separating my analyte from the matrix components (assuming the analyte/matrix have been dissolved or the analyte has been extracted into solution from the solid matrix). Gas chromatography is not suitable because nanoparticles are nonvolatile and cannot be vaporized while keeping their molecular structure intact. TiO2 and ZnO nanoparticles are inorganic solids that are not polar, hydrophobic/hydrophilic, ionic, or chiral, so the other types of chromatography (reverse-phase, HILIC, ion-exchange, and chiral, respectively) would not be suitable. They also do not have any specific affinity with a particular compound, so affinity chromatography is not suitable. Nanoparticles can be functionalized with organic molecules that give them some of these properties, but this is not relevant to my analytical problem.

2. Size-exclusion chromatography is the best/only chromatographic technique for my analytical problem, although other separation techniques, such as field-flow fractionation, can be used for nanoparticles (4). SEC is suitable for high molecular weight analytes and separates them by size rather than molecular properties. Separating nanoparticles by size would also be useful to determine whether smaller nanoparticles penetrate further into skin than larger ones.

3. The column used by Krueger et al to separate CdSe nanoparticles was an Agilent PLgel Individual Pore Size GPC/SEC Column (13).

Stationary phase: porous cross-linked polystyrene
Particle size: 5 micrometers
Pore size: 100 nanometers
Molecular weight range: 500-60,000 amu
Column diameter: 7.5 mm
Column length: 300 mm
pH stability: pH 1-14
Maximum temperature: 150 C
Typical pressure (for a similar column): 30 bar per 300 mm at flow rate 1 mL/min
Maximum flow rate (similar column): 1.5 mL/min
Maximum pressure (similar column): 150 bar
Agilent catalog number: PL1110-6530
Price: $1525

Product specifications from literature source 13 and Agilent:

4. The mobile phase used by Krueger for CdSe nanoparticles was toluene (13). The polarity of the solvent is not as important for size-exclusion chromatography as for some other types of chromatography, since this technique does not rely on interactions between polar or nonpolar molecules. Toluene would probably work well, assuming ZnO and TiO2 nanoparticles also form a stable suspension in it.

5. I would use ICP-MS, which would quantify the amount of Ti or Zn. This is the most suitable technique for quantification of my analyte, since the absorbance and other optical properties of nanoparticles depend on their size and shape. The number of nanoparticles in the sample could be calculated from the amount of Ti or Zn present, the diameter of the particles (measured with TEM or SEM), and the known stoichiometry and density of the materials.

In one experiment with sunscreen, Ti was detected at levels of 1-20 ug/g tissue in the dermis and 1-10 mg/g tissue in the epidermis, using ICP-MS on microwave-dissolved skin samples (6). Depending on my sample preparation for chromatography, my analytes could be more or less concentrated, but it seems reasonable that they would still be within the range of concentrations detectable by ICP-MS.

Literature source 4 was originally cited in Blog 2; source 6 was cited in Blog 4.
13. Krueger, K. M. et al. Characterization of Nanocrystalline CdSe by Size Exclusion Chromatography. Anal. Chem. 2005, 77, 3511-3515.

Good answers.

Blog 8: Sample Preparation Procedures

Skin sample preparation methods are very different for imaging and quantitative (atomic spectrometry or MS) methods; I looked at TEM and ICP-MS sample preparations as examples of these two types of techniques. Also, it is unlikely that I will use a chromatographic separation for my analytical problem.

ICP-MS sample preparation: (6)
1. Microwave dissolution: A mixture of HNO3 and HF (4:1 v/v) will be added to the skin sample and microwaved for 35 minutes (300 W power, 200 C, 220 psi).
2. Dilution: The dissolved sample will be diluted with 2% HNO3.
3. Internal standard addition: An internal standard (literature source used yttrium) for ICP-MS will be added to the sample.

TEM sample preparation: (6,7)
1. Fixation: The sample will be placed in a fixative solution (e.g. formaldehyde and glutaraldehyde in cacodylate buffer, or similar solutions commercially available as Karnovsky’s fixative or Trump’s fixative) for several hours.
2. Rinsing: The sample will be rinsed several times with phosphate, cacodylate, and/or acetate buffer.
3. Dehydration: Water will be removed from the sample by rinsing with increasing concentrations of ethanol.
4. Embedding and slicing: The sample will be placed in an epoxy resin mixture and hardened in an oven, then the resin block will be trimmed and slices (90 nm thick) will be cut from areas of interest.
5. Mounting: The resin-embedded sample will be mounted on a copper grid for TEM.
6. Staining: The sample can be stained with lead citrate or uranyl acetate. This improves contrast and increases resolution, but can cause interference in the signal.
7. Carbon coating: Finally, the sample will be vacuum-coated with a layer of carbon to prevent heat and charge buildup from damaging it during TEM.

Both literature sources were previously cited in Blog 4.

Blog 7, regraded, No points were taken off.

Blog 6. Good answers and logic.
Blog 7. Example of mass spectrum? (-0.1 Pt)

Blog 7: Atomic and mass spectrometries

3. Atomic spectrometry

The analytes are Zn and Ti atoms, from ZnO and TiO2 nanoparticles. Emission wavelengths include 213.857, 202.548, 206.200, and 334.501 nm for Zn, and 334.940, 336.121, 337.279, and 368.519 nm for Ti (11). I would probably not use the wavelengths at 334 nm in order to avoid overlap if both analytes are present.

The type of spectrometry is ICP-AES. ICP is useful because it generates atoms of all elements at once, so Zn and Ti could be analyzed simultaneously. It has been used before to detect these metals in skin and tissue samples (6,7 from Blog 4). However, it would be important to consider background levels of naturally occurring Zn in the body.

4. Mass spectrometry

The average mass of Ti (from the periodic table) is 47.88 g/mol. Ti primarily occurs as the 48Ti isotope, which has an exact mass of 47.9479463 amu (73.72% abundance). The next most abundant isotopes are 46Ti (45.9526316 amu, 8.25%) and 47Ti (46.9517631 amu, 7.44%). (12)

The average mass of Zn is 65.37 g/mol. Zn primarily occurs as the 64Zn isotope, which has an exact mass of 63.9291422 amu (48.268% abundance), as well as 66Zn (65.9260334 amu, 27.975%) and 68Zn (67.9248442 amu, 19.024%). (12)

The type of mass spectrometry I would use is TOF-SIMS (time-of-flight secondary ion mass spectrometry). This technique can be used for the detection of metal in skin samples. Chemical ionization is used to sputter a spot on the sample and erode ions from the sample surface. The ions are then analyzed by a TOF mass analyzer. TOF-SIMS can be used to generate images that relate Zn or Ti concentration to the position on the sample (7). Alternatively, ICP-MS can be used, probably with a quadrupole mass analyzer, to quantify Zn or Ti concentration in tissue samples (6).

11. Zachariadis, G.A. and Sahanidou, E. Multi-element method for determination of trace elements in sunscreens by ICP-AES. J Pharm. Biomed. Anal. 2009, 50, 342-348.
12. CRC Handbook of Chemistry and Physics, 92nd ed. Accessed online via CRCnetBASE at http://www.hbcpnetbase.com.

Good answers for BLOGS 4 and 5.

Note: After further researching the types of studies relevant to this topic, I would like to revise my analytical problem and only consider the penetration of nanoparticles into skin, not blood. Skin (solid) and blood (liquid) are very different matrices and are analyzed by very different techniques. It seems more relevant to consider them as two separate analytical problems. Also, if nanoparticles don't penetrate very far into skin, there's probably no point in looking for them in blood. From this point on, I will focus on detecting nanoparticles in a skin matrix.

Blog 5: Fluorescence techniques
ZnO and TiO2 display fluorescence, but by a different mechanism than the molecular excitations discussed in class. The fluorescence of these semiconductor materials corresponds to electrons moving between the valence and conduction bands.

2. ZnO: excitation 320 nm, emission 385 nm for nanoparticles in human skin (9)
It was observed that excitation at 320 nm also caused fluorescence of molecules naturally present in skin, such as NADH and NADPH. The solution was to use two-photon excitation of ZnO with a 740 nm laser, which reduced background fluorescence and could also penetrate further into skin.

TiO2: excitation 248 nm, emission 420-500 nm for colloidal nanoparticles in water, depending on crystal structure and preparation (10)
TiO2 is an indirect band gap semiconductor, so its fluorescence mechanism is not as straightforward as the mechanism for ZnO and it produces broader, less consistent spectra. Fluorescence has mostly been used with dye-conjugated TiO2 nanoparticles or to monitor its photocatalytic activity by measuring reaction products, but neither of these applications is relevant to my analytical problem.

5. The best use of fluorescence for my analytical problem, considering the skin matrix, would probably be similar to the fluorescence microscopy setup used by Zvyagin and coworkers for imaging ZnO in human skin. The sample was excited by a pulsed laser at 740 nm, and emitted light was collected at both 380 nm (narrow filter, ZnO fluorescence) and up to 700 nm (broad filter, skin autofluorescence background). Scanning the focal spot across the sample allows an image to be generated. For my analytical problem, fluorescence would be more useful for imaging nanoparticles and determining their location in skin, rather than quantifying their concentration.

9. Zvyagin, A. et al. Imaging of zinc oxide nanoparticle penetration in human skin in vitro and in vivo. J. Biomed. Opt. 2008, 13, 064031.
10. Monticone, S. et al. Quantum size effect in TiO2 nanoparticles: does it exist? Appl. Surf. Sci. 2000, 162, 565-570.

Blog 4: Studies needed to investigate analytical problem

Hypothesis: Nanoparticles from sunscreen only penetrate the top layers of human skin and are not found in blood after typical use.

1. Studies

A. Characterization of nanoparticles in sunscreen
Procure several different types of commercially available sunscreen containing ZnO and/or TiO2 nanoparticles, as well as sunscreens containing bulk ZnO and TiO2. Determine the material, size, coating, and concentration of the particles in each sunscreen. This study is important to a full understanding of the problem because there are currently no standards for particle size, and manufacturers are not required to report this information on product labels (6). Conclusions about the outcomes of human exposure to nanoparticles cannot be drawn without knowing what the nanoparticles are.

B. Exposure of humans to nanoparticles
Recruit a diverse sample of human subjects and instruct them to apply sunscreen multiple times a day for several days. Compare sunscreens containing nanoparticles, sunscreens containing bulk ZnO and TiO2, and no sunscreen application. If possible, collect skin and blood samples each day (note: I need to do more research into analysis of skin to determine how samples are collected and if removing skin from human subjects is feasible). Determine the nanoparticle content in different layers of skin, preferably using both an imaging technique and a technique for quantifying ZnO and TiO2 concentration. Determine the nanoparticle and metal ion concentration in blood.

C. Controlled study on pigs
If the results from the study on humans are inconsistent with my hypothesis and/or the literature on this subject, I would conduct a more controlled study on pigs. Pig skin is often used as a model for human skin, and has been used in very similar studies (6,7). The same types of sunscreens and frequency of application as in the human study would be used. The same samples would be collected, and they would be analyzed using the same techniques. The benefit of this study is that most conditions (amount of sunscreen applied, frequency of application, locations of application, exposure to sun) can be more easily controlled than in a study of humans. This study would hopefully confirm the results of the human study; if not, it would indicate that more controlled and/or extensive studies should be conducted on humans.

2. Estimate of analyte levels in samples
~5 mg/g tissue in the epidermis, ~10 ug/g tissue in the dermis (possibly due to incomplete separation from epidermis, so actual amounts may be smaller) (6)
~3 ug/L in the bloodstream, in the form of ions (calculated based on measurement of total Zn in blood and using isotope ratios to determine how much what came from nanoparticles) (8)

6. Sadrieh, N. et al. Lack of Significant Dermal Penetration of Titanium Dioxide from Sunscreen Formulations Containing Nano- and Submicron-Size TiO2 Particles. Toxicol. Sci. 2010, 115, 156-166.
7. Monteiro-Riviere, N. A. et al. Safety Evaluation of Sunscreen Formulations Containing Titanium Dioxide and Zinc Oxide Nanoparticles in UVB Sunburned Skin: An In Vitro and In Vivo Study. Toxicol. Sci. 2011, 123, 264-280.
8. Gulson, B. et al. Small Amounts of Zinc from Zinc Oxide Particles in Sunscreens Applied Outdoors Are Absorbed through Human Skin. Toxicol. Sci. 2010, 118, 140-149.

BLOG 2 - The last paragraph is not necessary. Remove. Save elsewhere because will need this information later. Once you have made the change notify Chad so that he releases your grade.

BLOG 3 - Parts (b) and (c)? (-0.5 pt)

Nate Vetter – Nanoparticles Accumulate in the Food Chain
This analytical problem is similar to mine. Both involve the use of commercially available nanoparticles and their possible health effects on living organisms. Heidi’s problem is more geared toward humans where as my problem could eventually lead to humans depending on the outcome. Both analytical problems involve the use of uv-vis to characterize the analytes.

I think my analytical problem is similar to Heidi Nelson's analytical problem because we both focuses on the nanosite of titanium dioxide to detect what is the rooted cause of the problem. Also I think we both will rely on UV-Vis spectrometery and then to use other spectroscopy.

Great posting!