March 8, 2007

Synthesis, Characterization, and Bioavailability in Rats of Ferric Phosphate Nanoparticles1

By Rohner, Fabian; Ernst, Frank O; Arnold, Myrtha; Hilbe, Monika; Et al


Particle size is a determinant of iron (Fe) absorption from poorly soluble Fe compounds. Decreasing the particle size of metallic Fe and ferric pyrophosphate added to foods increases Fe absorption. The aim of this study was to develop and characterize nanoparticles of FePO^sub 4^ and determine their bioavailability and potential toxicity in rats. Amorphous FePO^sub 4^ nanopowders with spherical structure were synthesized by flame spray pyrolysis (FSP). The nanopowders were characterized and compared with commercially available FePO^sub 4^ and FeSO^sub 4^, including measurements of specific surface area (SSA), structure by transmission electron microscopy, in vitro solubility at pH 1 and 2, and relative bioavailability value (RBV) to FeSO^sub 4^ in rats using the hemoglobin repletion method. In the latter, the potential toxicity after Fe repletion was assessed by histological examination and measurement of thiobarbituric acid reactive substances (TBARS). The commercial FePO^sub 4^ and the 2 FePO^sub 4^ produced by FSP (mean particle sizes, 30.5 and 10.7 nm) had the following characteristics: SSA: 32.6, 68.6, 194.7 m^sup 2^/g; in vitro solubility after 30 min at pH 1: 73, 79, and 85% of FeSO^sub 4^; and RBV: 61, 70, and 96%, respectively. In the histological examinations and TBARS analysis, there were no indications of toxicity. In conclusion, nanoparticles of FePO^sub 4^ have a solubility and RBV not significantly different from FeSO^sub 4^. Reducing poorly soluble Fe compounds to nanoscale may increase their value for human nutrition. J. Nutr. 137: 614- 619, 2007.


Iron deficiency anemia is a major global public health problem (1). Food fortification with iron (Fe) can be an effective strategy to control iron deficiency anemia, but adding Fe to foods can be problematic (2). Water-soluble, highly bioavailable Fe compounds often cause adverse organoleptic changes, whereas low solubility Fe compounds, although more stable in foods, tend to have low bioavailability. Particle size can be an important determinant of Fe absorption from poorly soluble Fe compounds in foods. Decreasing the particle size of metallic Fe powders by 50-60% to a mean particle size (MPS)6 of 7-10 m, as determined from dynamic light scattering (d^sub LS^), has been reported to increase Fe absorption by 50% in rats (3,4). In humans, Fe absorption from hydrogen-reduced Fe with an MPS of 5 and 10 m was comparable to that of ferrous sulfate (5). Reducing the particle size of ferric pyrophosphate by conventional grinding from a MPS of ≥20 m to 2-3 m resulted in a nonsignificant increase in its relative bioavailability value (RBV) to ferrous sulfate (6). However, smaller ferric pyrophosphate particles, with an MPS of 0.3-0.5 m in aqueous solution with emulsifiers preventing agglomeration, have an RBV of ~95% (7,8). These studies suggest reducing the particle size of low solubility Fe compounds to submicron size may be promising for food fortification, because they may cause fewer sensory changes yet be highly bioavailable.

Nanotechnology deals with materials and systems of characteristic scale below 100 nm to exploit novel properties and phenomena (9). Potential medical/nutritional applications for nanomaterials include new systems that may allow targeted delivery of substances, as well as enhanced permeability and increased retention (10,11). Recent research on the gastrointestinal absorption of nanoparticles has focused on enhancing the absorption of drugs, vaccines, and nutrients that are either degraded by the digestive process or poorly absorbed (12,13). However, there are toxicity concerns with nanoparticles. As most studies in this area have focused on airborne nanoparticles, there are only limited data on orally ingested trace element nanoparticles (14,15) focusing on selenium, copper, and zinc (16-18). Nanoparticles may be taken up by persorption and/or absorbed by gut-associated lymphoid tissue and pass through the mesenteric lymphatics to the venous circulation (19).

Ferric phosphate (FePO^sub 4^) is a white-colored, poorly soluble Fe compound of little nutritional value due to its low bioavailability (20). In this study, we used flame technology to produce FePO^sub 4^ nanoparticles of various sizes. This is a fast, dry, and versatile process for synthesis of nanostructured commodities such as carbon black, fumed silica, alumina, etc. (21). Flame spray pytolysis (FSP) is a scalable process (22) allowing for production of tailor-made particles with high specific surface area (SSA) and well-defined chemical composition (23). We aimed to develop and characterize nanoparticles of FePO^sub 4^ and determine the effects of particle size reduction into the nanorange on bioavailability and safety of these particles in rats.

Materials and Methods

Commercially available compounds. In this study, 4 Fe compounds were analyzed. Two of them were commercially available FePO^sub 4^ and ferrous sulfate hydrate (FeSO^sub 4^.H2O), obtained from Dr. Paul Lohmann GmbH KG, Emmerthal, Germany (Art. nos. 3043355 and 3590548, respectively).

Nanoparticle synthesis. FePO^sub 4^ nanoparricles were made by FSP (24). The precursors, Fe(III)-acerylacetonate and tri- butylphosphate (reagent grade, Fluka), were dissolved in xylene (Riedel-de-Haen, >96%, dried over molecular sieves) at an Fe and phosphorus ion concentration of 0.2 mol/L each. This solution was metered into the reactor nozzle by a syringe pump (Inotech R232) at a rate of 3-7 mL/min and dispersed by 3-8 L/min O2 (PanGas, purity 99.95%) into fine droplets by a gas-assist nozzle; the pressure drop at the nozzle tip was 1.5 bar. The spray was ignited by a premixed methane/oxygen flame ring surrounding the spray capillary at a radius of 6 mm with a spacing of 0.15 mm (25). The premixed flame was fed by 1.13 L/min CH^sub 4^ (PanGas, purity 99.5%) and 2.40 L/ min O2 (PanGas, purity 99.95%). An additional oxygen sheath flow of 5 L/min was fed through a sinter metal ring of 8-mm width and 9-mm i.d., surrounding the supporting flame to ensure complete conversion of the reactants. The rates of the liquid feed and dispersion oxygen flow were varied to select the desired particle characteristics (24). The gas flows were monitored by calibrated mass flow controllers (Bronkhorst). Precursor evaporation within the liquid feed lines and nozzle overheating was prevented by water cooling the reactor. Using a vacuum pump, product particles were collected on a Teflon-supported Teflon membrane filter (BHA Technologies, 1TMTF700WHT, Muemliswil) placed in a water-cooled holder 50-65 cm above the nozzle, keeping the off-gas temperature below 200C.

Powder characterization. The powder SSA was determined by the Brunauer-Emmett-Teller (BET) method from a 5-point nitrogen adsorption isotherm at 77 K in. the relative pressure range (p/^sub p^0) = 0.05 to 0.25 (Tristar, Micromeritics Instruments). Prior to analysis, samples were outgassed at 150C for 1 h. Assuming dense spherical particles, the particle diameter as calculated from SSA measurements (d^sub BET^) was calculated according to d^sub BET^ = 6/ (powder SSA). For transmission electron microscopy (TEM) investigation, particles were dispersed in ethanol and deposited onto a carbon foil supported on a copper grid. TEM investigations were performed on a CM30ST microscope (FEl; LaB^sub 6^ cathode, operated at 300 kV, point resolution ~2[Angstrom]). Selected area electron diffraction (SAED) was used to verify the amorphous state of the synthesized powder.

The stochiometric composition of the FePO^sub 4^ particles was analyzed using inductively coupled plasma mass spectrometry (ICP- MS). A total of 200 mg of sample was dissolved in 5% nitric acid and analyzed on a sector field ICP-MS (Element 2, Finigan-MAT). The instrument was equipped with a perfluoroalkoxy (PFA)-screw top (ST) microflow nebulizer operated at a sample uptake rate of 300 L/min and a PFA spray chamber, both from Elemental Scientific. The mass spectrometer was operated in a medium resolution mode (m/Δm = 4000) to separate molecular interferences ^sup 40^Ar^sup 16^O^sup +^ and ^sup 15^N^sup 16^O^sup +^ from the monitored ions, ^sup 56^Fe and ^sup 31^P, respectively. The electric scan acquisition mode was used and 9 scans were performed per sample. The optimization was restricted to adjustments of sample (0.8 L/min), make-up (0.3 L/ min), and auxiliary (0.6 L/min) gas flow rates to obtain a stable response at a maximum signal, and 1.3 kW rf-power was applied at a cool gas flow rate of 16 L/min. The obtained ratios were subsequently converted into molecular formulae assuming the existence of hydrates.

Raman spectroscopy was performed on FePO^sub 4^ panicles using a Renishaw InVia Reflex Raman system equipped with a 514-nm diode (solid state, 25 mW) laser as an excitation source focused in a microscope (Leica, city magnification 50). Three spectra were recorded (60 s) on a charge-coupled device camera after diffraction (1200 lines . mm^sup -1^) using 0.25-mW laser energy to avoid thermal alteration. The spectra obtained were compared with those obtained from the commercially available FePO^sub 4^.

To characterize the Fe species (Fe^sup 2+^/Fe^su\p 3+^) in the FePO^sub 4^ particles, they were dissolved in sulfuric acid and underwent reduction-oxidation using cerium as reducing agent and ferroin (1,10-phenanthroline) as indicator (26). This method allows visual detection of the occurrence of ferrous Fe >5% of the total Fe content. After dissolution of Fe powders in hydrochloric acid (32%), their Fe content was measured by atomic absorption spectrometry (SpectrAA-240FS; Varian).

The in vitro solubility of the Fe compounds was tested following Swain et al. (27). A compound containing 20 mg Fe was added to 250 mL aqueous solution of 0.1 and 0.01 mol/L hydrochloric acid, corresponding to pH 1 and 2, respectively, and mixed in an orbital shaker at 150 rpm and 37C (Aqua shaker, Adolf Kuhner). The percentage of dissolved Fe was assessed after 5, 15, 30, 45, and 60 min in a 1.5-mL solution aliquot. After centrifugation (11,600 g; 4 min), the Fe content of the supernatant solution was measured by atomic absorption spectroscopy (AAS) with external calibration.

Rats and diets. Ethical approval for the study was granted by the Veterinary Office of the Canton of Zurich's Department of Health. The bioavailability of the Fe compounds was determined by the hemoglobin (Hb) repletion method (20,28,29). Two levels of dietary Fe were used for each compound. Male Sprague-Dawley rats (n = 115; Charles River) 21 3 d old were housed individually in stainless steel cages with grated stainless steel floors. The rats were kept at 23.6 0.5C and a relative humidity of 51.3 6.8%, with a 12-h- light/-dark cycle. AH rats were handled daily to reduce stress at blood collection and killing. Body weight was measured 3 times/wk. Rats consumed Millipore water (Milli-Q UF Plus, Millipore) ad libitum. The diets were prepared by Dyets following the AIN-93G purified rodents guidelines (30). The rats consumed an Fe-deficient (Fe-def) diet (Fe content, 2.5 mg/kg, as measured by AAS) ad libitum for 24 d. After this depletion period, rats were weighed and blood was collected by tail incision (31), with precautions taken to avoid hemodilution. To increase tail vein vasodilation, the tail was wrapped in a warming towel. The blood was collected into heparincoated capillaries for immediate Hb determination and into EDTA-coated capillary tubes (Microvette 300, Sarstedt) for further analysis of plasma. Blood from these containers was centcifuged (3000 g; 8 min at 4C) and plasma for thiobarbituric acid reactive substances (TBARS) measurement was separated and stored at -80C.

Rats with an Hb concentration of 35 4 g/L (range 28-46 g/L) after depletion were randomly assigned to 9 groups of 9-12 rats. The rats of each group consumed the same Fe-def diet fortified with 1 of the 3 FePO^sub 4^ compounds (each at intended levels of 10 or 20 mg Fe/kg diet), ferrous sulfate (FeSO^sub 4^.H2O; at 10 or 20 mg Fe/kg diet), or no added Fe (2.5 mg Fe/kg diet; Fe-def) for 15 d ad libitum. Other than their Fe concentrations, the diets were equivalent and conformed to the recommendations for AIN-93 purified diets (30). The Fe concentration of all diets was verified by AAS (SpecrAA-240Zwith GTA-120 Graphite Tube Atomizer, Varian Techtron). To serve as Fe-sufficient controls, 3 rats received a regular rodent diet (Kliba) throughout the experimental period (Fe-sufficient). These rats were used for comparisons in the histological assessment. Individual food consumption was recorded daily throughout the repletion period. After the repletion period, rats were weighed, and blood was collected by tail incision and processed as described above, with plasma for TBARS measurement stored at -80C. The rats were then killed using CO2.

Laboratory analysis. Hb concentration was measured in triplicate in whole blood with a commercial kit (D5941; Sigma) using the cyanmethemoglobin method (32) and commercially available control material (Digitana AG). TBARS were measured in plasma in duplicate with a commercially available test (TBARS assay kit, ZeptoMetrix); normal plasma concentrations are

TABLE 1 Compound characteristics: SSA, calculated MPS (d^sub BET^), physical structure, stochiometry (chemical composition), and in vitro solubility after 15 and 30 min in 0.1 mol/L HCI (SOL) and RBV in rats1

Histological examination. From 3 of the rats in each of the following 6 groups: control (Fe-suffieicnt), Fe-def, the 3 FePO^sub 4^ compounds fed at 20 mgFe/kg, and the FeSO^sub 4^ fed at 20 mg He/ kg (total n = 18), tissue samples of the stomach, duodenum, jejunum, ileum, colon, liver, spleen, kidney, pancreas, lymphatic tissue, and sternum were excised immediately after killing. For light microscopy, tissues were fixed by immersion in 4% buffered formaldehyde, dehydrated with xylene and a descending alcohol row (Tissue Tek VIP), paraffin embedded, and subsequently stained with hematoxylin-eosin, Prussian Blue for detection of Fe^sup 3+^, and Turnbull Blue for detection of Fe^sup 2+^. TEM (CM 10 Philips) was used to examine sections of duodenal mucosa from a single rat from each group (n = 6). Tissues were fixed in 2.5% glutaraldehyde and embedded in epoxy before further processing into ultrathin sections. The veterinary pathologists and microscopists were unaware of the group assignment.

Statistical analysis. Data processing and analyses were done using SPLUS-2000 (Release 3, Insightful Corporation), SPSS (version 13.0; SPSS) and EXCEL (Enterprise Edition; Microsoft). Values in the text are means SD. Using the slope-ratio method, the hioavailability (RBV) of each Fe compound relative to FeSO^sub 4^ was calculated by comparing the change in Hb [g/(L.15 d)| with the measured Fe intake (g/d) (33,34). The slope of the responses for each dietary Fe compound was calculated by using a common-intercept multiple linear regression model with the Fe-def group serving as the blank. Linearity of the regression curves was determined for each Fe compound separately and tests were conducted to determine whether the mean of the blank differed significantly from the common intercept for the 4 Fe compounds. Tukey's method was applied to test whether the slopes of the 3 FePO^sub 4^ compounds were significantly different from that of FeSO^sub 4^ and from each other. Using Fieller's method (35), [95% CI] for the RBV to ferrous sulfate were obtained. To compare means between the treatment groups, 1-way ANOVA was done with post-hoc t tests adjusted for multiple comparisons (Bonferroni). Independent sample t tests were used to compare the in vitro solubility results, and adjusted for multiple comparisons (Bonferroni). Percentages were compared using chi-square tests. Differences were considered significant at P


Material characterization. The SSA of the 3 FePO^sub 4^ compounds and their calculated d^sub BET^ are shown in Table 1. The 2 FSP- made FePO^sub 4^, medium and small particles, were dense spherical particles and the d^sub BET^ matched well the observation from TEM (Fig. 1B,C). The TEM image of the commercial powder, FePO^sub 4^ large particle (Fig. 1A), shows irregular and highly porous particles. For all 3 FePO^sub 4^ compounds, the SAED images were characteristic of an amorphous substance (Fig. 1). Handling characteristics of the fine powders were similar to commercially available small particle size Fe compounds, such as ground micronized ferric pyrophosphate (Dr. Paul Lohmann GmbH KG, Emmerthal, Germany).

For the FSP-made FePO^sub 4^ medium and small particles, ICP-MS demonstrated a Fe:P ratio of 1.93 as opposed to the expected 1.80 for anhydrous FePO^sub 4^. Including hydrates in the ratio calculation, best fit with the expected ratio was obtained with FePO^sub 4^.2H^sub 2^O (ratio = 1.80). Raman spectroscopy confirmed the presence of Fe phosphate (data not shown). Using the Fe speciation assay, there were no detectable ferrous ions, indicating that >95% of the Fe was in the ferric state (data not shown). AAS analysis revealed Fe contents of the FePO^sub 4^ large, medium, and small particles of 25.6 0.4%, 33.8 0.8%, and 33.2 0.5%, respectively.

In vitro solubility. In the tests of in vitro solubility, all 3 FePO^sub 4^ compounds were very poorly soluble at pH 2; solubility was

RBV. The results of the Hh repletion study, including diet fortification level, rat number per group, daily Fe intake, body weight gain, and Hb change over the repletion period are shown in Table 2. Dose-response curves were calculated based on daily Fe intake (Fig. 3). In this model, the regression lines for the 4 Fe compounds did not significantly deviate from linearity and the mean of the blank (circles) was not significantly different from the common intercept (data not shown). The RBV of FePO^sub 4^ large and FePO^sub 4^ medium particles did not differ from each other, but both were lower than the RBV of the FePO^sub 4^ small particle (P

Figure 1 TEM and SAED (insels) images of the 3 FePO^sub 4^ compounds: (A) FePO^sub 4^ large particle. (B) FePO^sub 4^ medium particle, and (C) FePO^sub 4^ small particle. The 2 compounds made by FSP, medium and small FePO^sub 4^ particles, were dense and spherical (Fig. 1B,C). The FePO^sub 4^ large particle (Fig. 1A) exhibited irregular and highly porous particles. For all 3 compounds, the SAED images were characteristic of an amorphou\s substance.

Figure 2 In vitro solubility test of the 3 FePO^sub 4^ compounds and FeSO^sub 4^ at pH 1. Values are means SD, n = 3. Values marked with *, **, *** are different from the reference compound, FeSO^sub 4^, P

Histology and TBARS. On the hematoxylin-eosin stained sections, histologie changes consistent with hypoxic damage to the liver were found in 3 rats from the Fe-def group. There were no visible inflammatory changes or other adverse findings in the tissues sampled. No detectable Fe^sup 2t^ was seen in the Tumbull Blue stains. In the Prussian Blue staining for Fe^sup 3+^, only in 2 of the rats receiving the 20 mg Fe/kg diet as FeSO^sub 4^, the reference material, were small amounts of Fe detected in the liver, gastric mucosa, and kidney. In the duodenum sections examined by TEM, no paniculate material consistent with Fe particles was seen.

Plasma TBARS did not differ among the groups at the end of the repletion period (Table 2) and all values were well within the reference range for the assay.


To our knowledge, this is the first study investigating the potential nutritional value of Fe-containing nanoparticles. The FePO^sub 4^ small particles, with a d^sub BET^ 10.7 nm, demonstrated in vitro solubility and in vivo RBV equivalent to commercial FeSO^sub 4^. Moreover, comparing the nanoparticulate FePO4 to commercial FeSO^sub 4^, there was no evidence of potential toxicity in the histologie examinations or the TBARS analyses. This suggests that FePO^sub 4^, considered a poorly soluble compound with a low RBV, when reduced to nanoparticle size, shows performance characteristics similar to FeSO^sub 4^, the reference Fe compound for food fortification for humans. Although sensory aspects were not considered in this study, compared with FeSO^sub 4^, the FePO^sub 4^ nanoparticles may possibly produce fewer adverse organoleptic changes in food vehicles, particularly in color-sensitive foods, such as rice, salt, milk-based drinks, and highly refined cereal flours. However, the sensory characteristics of the Fe nanoparticles needs further investigation.

TABLE 2 Fe fortification level, fortified Fe intake, body weight gain, change in Hb in Fe-depleted rats during the 15-d repletion period, and plasma TBARS at d 15(1)

Few studies have investigated the potential of FePO^sub 4^ for food fortification (20,36,37). Despite good sensory characteristics (Bright color, only minor off-flavors in foods), FePO^sub 4^ is generally considered to have little nutritional value due to its low hioavailability. In these studies, no consideration was given to whether the FePO^sub 4^ was in the crystalline or amorphous state, although this structural characteristic was shown to significantly affect FePO^sub 4^ hioavailability (38). Compared with previous studies, the higher relative bioavailability of commercial FePO^sub 4^ used in the present study may have been, at least partially, caused by its amorphous state, its irregular porous structure, and its high surface area. It was reported that the RBV of FePO^sub 4^ varies between 6-46% (2) and in a study comparing crystalline with amorphous FePO^sub 4^ in humans (36), higher bioavailability was found for the latter (20 vs. 28%).

Two previous studies have demonstrated that metallic Fe particles of m (39) and nm (40) size may cross into the circulation through paracellular uptake through tight junctions. Inert nanoparticles may also be absorbed through Peyer's patches of the small intestine, passing through the mesenteric lymphatics to the liver and spleen (41). Particles that are smaller, hydrophilic, and positively charged are more readily absorbed by this process (42,43).

This study does not suggest absorption of the FePO^sub 4^ nanoparticles via paracellular uptake or the lymphatic system, as there was no visible or stainahle Fe in the mesenteric lymphatics in the histologic sections. Considering the high solubility of the FePO^sub 4^ nanoparticles, they were most likely dissolved during digestion in the intestinal lumen and absorbed through the usual receptor-mediated pathway of nonheme Fe absorption, i.e. uptake of Fe by the divalent metal transporter 1 on the duodenal brush border (44).

To assess potential oxidative stress in the rats during Fe repletion, we measured the production of TBARS, a by-product of reactive oxygen processes. Previous in vitro and in vivo studies on the toxicology of airborne nanoparticles demonstrated increased production of reactive oxygen species, which has been attributed to preferential mobilization of the nanoparticles to mitochondria and/ or redox active organelles (15), their increased surface area (45), and/or the presence of free Fe (46). However, given orally in this study, the Fe nanoparticles did not appear to promote oxidation; there were no differences in TBARS production among the groups administered the various Fe compounds.

Figure 3 Dose-response curves for the Hb repletion assay m depleted rats consuming a Fe-def diet or the Fe-def diet fortified in graded concentrations with FeSO^sub 4^, FePO^sub 4^ small particle, FePO^sub 4^ medium particle, or FePO^sub 4^ large particle for 15 d. Regression (mes were calculated on daily Fe intake (g/d) and change in Hb concentration [g/(L.15 d)]. Values are shown individually, n = 9-12/ group.

Data on the nutritional and/or toxicological effects of other nanoscale trace elements is limited. Nanosized particles of selenium (30-60 nm) fed to rats had a bioavailability comparable with sodium selenite (17) and lower subchronic toxicity (47). In contrast, in mice, the acute oral toxicity of nanoscale zinc (58 nm) and copper (24 nm) were higher than equivalent amounts of microscale zinc (16,18). Although this study was not specifically designed to investigate the toxicity of Fe compound nanoparticles, feeding 150- 370 g Fe/d for 15 d to weanling rats did not induce measurable histologic or biochemical adverse effects. Further studies are needed to determine the acute and subchronic toxicity of Fe- containing nanoparticles.

The SSA represents the accessible surface area. Thus, if the d^sub BET^ derived from the SSA under the assumption of dense spheres is smaller than the diameter measured by light scattering, the particles are either porous or agglomerated. The FePO^sub 4^ large particles investigated in this study were porous and irregular (Fig. 1A) and the measured particle size (d^sub LS, 50^ = 2.5 /m), as well as the manufacturer's specifications for particle size based on sieve analysis (d^sub 50^ = 4.4 m), did not reflect the actual SSA (d^sub BET^ = 64.2 nm) of the material due to additional internal surface. Solubility is a surface-dependent phenomenon and thus determined by the SSA rather than the overall particle size as, for example, measured by light scattering. This may explain earlier studies where no significant differences in RBV were found despite an apparent reduction in particle size (d^sub LS^) of ferric pyrophosphate (3,4,6,48).

Flame technology is currently being used on a large scale to produce carbon black, fumed silica, and titania pigments (49). This established technology may thus be potentially competitive and cost effective for production of FePO^sub 4^ nanoparticles (or nanoparticles of other Fe compounds) for nutritional supplementation and/or food fortification. The results of this study suggest further research on the synthesis, efficacy, and safety of Fe-containing nanoparticles in nutrition would be valuable.


We thank Sabine Renggli, Isabelle Aeberli, Frank Bootz, Noemi Hernadfalvi, Matthias Hoppler, and Leah Standring (ETH, Zurich) for their technical assistance, Max Haldimann (The Ministry of Health, Bern, Switzerland) for the ICP-MS measurements. Frank Krumeich (ETH, Zurich) and Lisbeth Nufer (VetSuisse faculty) for the EM images of the Fe powders and the biological tissues, and Luciano Molinari (Children's Hospital, Zurich) for support in statistical analysis. We thank Paul Lohmann GmbH (Emmerthai, Germany) for providing the FePO4 and FeSO4 for the study at no cost, and BHA Technologies (Muemliswil, Switzerland) for supplying the Teflon filters at no cost.

0022-3166/07 $8.00 2007 American Society for Nutrition.

Manuscript received 18 October 2006. Initial review completed 29 November 2006. Revision accepted 19 December 2006.

1 Supported by the ETH Zurich, Switzerland.

6 Abbreviations used: MS, atomic absorption spectrometry; BET, Brunauer-Emmett-Teller method; d^sub BET^, particle diameter as calculated from SSA measurements; d^sub LS^, particle diameter as measured by light scattering; Fe-def, Fe-deficent; FSP, flame spray pyrolysis; Hb, hemoglobin; ICP-MS, inductively coupled plasma mass spectrometry; RBV, relative bioavailability value; MPS. mean particle size; SAED, selected area electron diffraction; SSA, specific surface area; TBARS, thiobarbituric acid reactive substance; TEM, transmission electron microscopy.

Literature Cited

1. WHO/UNICEF/UNU. Iron deficiency anaemia. Assessment, prevention and control. A guide for programme managers. Geneva: WHO; 2001.

2. Hurrell RF. Fortification: overcoming technical and practical barriers. J Nutr. 2002;132:S806-12.

3. Motzok I, Pennell MD, Davies MI, Ross HU. Effect of particle size on the biological availability of reduced iron. J Assoc Off Anal Chem. 1975; 58:99-10.1.

4. Verma RS, Motzok I, Chen SS, Rasper J, Ross HU. Effect of storage in flour and of particle size on the bioavailability of elemental iron powders for rats and humans. J Assoc Off Anal Chem. 1977;60:759-65.

5. Cook JD, Minnich V, Moore CV, Rasrnussen A, Bradley WB, Finch CA. Absorption of fortification iron in bread. Am J Clin Nutr. 1973;26: 861-72.

6. Wegmuller R, Zimmermann MB, Moretti D, Arnold M, Langhans W, Hurrell RF. Particle size reduction and encapsulation affect the bioavailabili\ty of ferric pyrophosphate in rats. J Nutr. 2004;134: 3301-4.

7. Fidler MC, Walczyk T, Davidsson L, Zeder C, Sakaguchi N, Juneja LR, Hurrell RF. A micronised, dispersihle ferric pyrophosphatc with high relative bioa variability in man, Br J Nutr. 2004;91:107-12.

8. Nanhu H, Nakata K, Sakaguchi N, Yamazaki Y, inventors. Mineral Composition. EU, European Patent EP 0870435A1. 1998.

9. Ross SA, Srinivas PR, Clifford AJ, Lee SC, Philbert MA, Hettich RL. New technologies for nutrition research. J Nutr. 2004;134:681-5.

10. Bogunia-Kubik K, Sugisaka M. From molecular biology to nanotechnology and nanomedicinc. Biosystems. 2002;65:123-38.

11. Emerich DH, Thanos CG. Nanotechnology and medicine. Expert Opin Biol Ther. 2003;3:655-63.

12. Hoffart V, Lamprecht A, Maincent P, Lecompte T, Vigneron C, Ubrich N. Oral bioavailability of a low molecular weight heparin using a polymeric delivery system. J Controlled Release. 2006;113:38- 42.

13. Florence AT. Issues in oral nanoparticle drug carrier uptake and targeting. J Drug Target. 2004; 12:65-70.

14. Nel A, Xia T, Madler L, Li N. Toxic potential of materials at the nanolevel. Science. 2006;311:622-7.

15. Oberdorster G, Oberdorster E, Oberdorster J. Nanotoxicology: an emerging discipline evolving from studies of ultrafine particles. Environ Health Perspect. 2005;113:823-39.

16. Chen Z, Meng H, Xing G, Chert C, Zhao Y, Jia G, Wang T, Yuan H, Ye C, et al. Acute toxicological effects of copper nanoparticles in vivo. Toxicol Lett. 2006;163:109-20.

17. Zhang JS, Gao XY, Zhang LD, Bao YP. Biological effects of a nano red elemental selenium. Biofactors. 2001;15:27-38.

18. Wang B, Feng WY, Wang TC, Jia G, Wang M, Shi JW, Zhang F, Zhau YL, Chai ZE Acute toxicity of nano- and micro-scale zinc powder in healthy adult mice. Toxicol Lett. 2006;161:115-23.

19. Jani P, Halbert GW, Langridge J, Florence AT. Nanoparticle uptake by the rat gastrointestinal mucosa: quantitation and particle size dependency. J Pharm Pharmacol. 1990;42:821-6.

20. Forbcs AL, Arnaud MJ, Chichester CO, Cook JD, Harrison BN, Hurrell RF, Kahn SG, Morris ER, Tanner JT, et al. Comparison of in vitro, animal, and clinical determinations of iron bioavailability: International Nutritional Anemia Consultative Group Task Force report on iron bioavailability. Am J Clin Nutr. 1989;49:225-38.

21. Pratsinis SE. Flame aerosol synthesis of ceramic powders. Prog Energy Combust Sci. 1998;24:197-219.

22. Mueller R, Madler L, Pratsinis SE. Nanoparticle synthesis at high production rates by flame spray pyrolysis. Chem Eng Sci. 2003;58: 1969-76.

23. Wegner K, Pratsinis SE. Scale-up of nanoparticle synthesis in diffusion flame reactors. Chem Eng Sci. 2003;58:4581-9.

24. Madler L, Pratsinis SE. Bismuth oxide nanoparticlcs by flame spray pyrolysis. J Am Ceram Soc. 2002;85:1713-8.

25. Madler L, Kammler HK, Mueller R, Pratsinis SE. Controlled synthesis of nanostructured particles by flame spray pyrolysis. Aerosol Sci. 2002; 33:369-89.

26. Moore JW, Anderson RC. Kinetics of the reaction of ceriumIV and arsenicIII ions. J Am Chem Soc. 1944;66:1476-9.

27. Swain JH, Newman SM, Hunt JR. Bioavailabiljry of elemental iron powders to rats is less than bakery-grade ferrous sulfate and predicted by iron solubility and panicle surface area. J Nutr. 2003;133:3546-52.

28. Fritz JC, Pla GW, Harrison BN, Clark GA. Collaborative study of rat hemoglobin repletion test for bioavailability of iron. J Assoc Off Anal Chem. 1974;57:513-7.

29. AOAC. AOAC official method 9743.1: bioavailability of iron: rat hemoglobin repletion bioassay. In: Cunniff P, editor. Official methods of analysis of AOAC International. 16th ed. Gaithersburg: AOAC International; 1997. p. 62-3.

30. Reeves PG, Nielsen FH, Fahey GC Jr. AIN-93 purified diets for laboratory rodents: final report of the American Institute of Nutrition ad hoc writing committee on the reformulation of the AIN- 76A rodent diet. J Nutr. 1993;123:1939-51.

31. Fluttert M, Dalm S, Oitzl MS. A refined method for sequential blood sampling by tail incision in rats. Lab Anim. 2000;34:372-8.

32. Crosby WH, Munn Jl, Furth FW. Standardizing a method for clinical hemoglobinometry. US Armed Forces Med J. 1954;5:693-703.

33. Amine EK, Hegsied DM, Neff R. Biological estimation of available iron using chicks or rats. J Agric Food Chem. 1972;20:246- 51.

34. Finney DJ. Efficiency in slope ratio assays. Statistical method in biological assay. 3rd ed. London: Charles Griffin & Company Ltd.; 1978. p. 214-27.

35. Finney DJ. Slope ratio assays. Statistical method in biological assay. 3rd ed. London: Charles Griffin & Company Ltd.; 1978. p. 187-213.

36. Hallberg L, Rossander-Hulthen L, Gramatkovski E. Iron fortification of flour with a complex ferric orthophosphate. Am J Clin Nutr. 1989; 50:129-35.

37. Harrison BN, Pla GW, Clark GA, Fritz JC. Selection of iron sources for cereal enrichment. Cereal Chem. 1976;5 3:78-84.

38. Willis RB, Alien PR. Measurement of amorphous ferric phosphate to assess iron bioavailability in diets and diet ingredients. Analyst, 1999; 124:425-30.

39. Volkheimer G, Schulz FH, Lindenau A, Beitz U. Persorprion of metallic iron particles. Gut. 1969;10:32-3.

40. McCullough JS, Hodges GM, Dickson GR, Yarwood A, Carr KE. A morphological and microanalytical investigation into the uptake of paniculate iron across the gastrointestinal tract of rats. J Submicrosc Cytol Pathol. 1995;27:119-24.

41. Hussain N, Jaitley V, Florence AT. Recent advances in the understanding of uptake of microparticulates across the gastrointestinal lymphatics. Adv Drug Deliv Rev. 2001;50:107-42.

42. Florence AT, Hillery AM, Hussain N, Jam PU. Factors affecting the oral uptake and translocation of polystyrene nanoparticles: histological and analytical evidence. J Drug Target. 1995;3:65-70.

43. Mathiowitz E, Jacob JS, Jong YS, Carino GP, Checkering DE, Chaturvedi P, Santos CA, Vijayaraghavan K, Montgomery S, et al. Biologically erodable microspheres as potential oral drug delivery systems. Nature. 1997;386:410-4.

44. Mackenzie B, Garrick MD. Iron imports. II. Iron uptake at the apical membrane in the intestine. Am J Physiol Gastrointest Liver Physiol. 2005;289:G981-6.

45. Joo SH, Feitz AJ, Sedlak DL, Waite TD. Quantification of the oxidizing capacity of nanoparticulate zero-valent iron. Environ Sci Technol. 2005; 39:1263-8.

46. Breuer W, Hershko C, Cabantchik ZI. The importance of nontransferrin bound iron in disorders of iron metabolism. Transfus Sci. 2000;23:185-92.

47. Jia X, Li N, Chen J. A subchronic toxicity study of elemental Nano-Se in Sprague-Dawley rats. Life Sci. 2005;76:1989-2003.

48. Fidler MC, Davidsson L, Zeder C, Walczyk T, Marti I, Hurrell RF. Effect of ascorbic acid and particle size on iron absorption from ferric pyrophosphatc in adult women. Int J Vitam Nutr Res. 2004;74:294-300.

49. Schultz H, Madler L, Pratsinis SE, Burtscher P, Mozner N. Transparent nanocomposites of radiopaque, flame-made Ta2O5/SiO2 particles in an acrylic matrix. Adv Functional Mater. 2005;15:830- 7.

Fabian Rohner,2 Frank O. Ernst,3 Myrtha Arnold,4 Monika Hilbe,5 Ralf Biebinger,2 Frank Ehrensperger,5 Sotiris E. Pratsinis,3 Wolfgang Langhans,4 Richard F. Hurrell,2 and Michael B, Zimmermann2*

2 Institute of Food Science and Nutrition, 3 Particle Technology Laboratory, and 4 Institute of Animal Science, ETH Zrich, Switzerland and 5 Institute of Veterinary Pathology, Vejsuisse Faculty, University of Zurich, Zurich 8092 Switzerland

* To whom correspondence should be addressed. E-mail: [email protected]

Copyright American Institute of Nutrition Mar 2007

(c) 2007 Journal of Nutrition, The. Provided by ProQuest Information and Learning. All rights Reserved.