Controlling the Self-Assembly Structure of Magnetic Nanoparticles and ...
Supporting Information
Controlling the Self-Assembly Structure of Magnetic Nanoparticles and Amphiphilic Blockcopolymers: From Micelles to Vesicles Robert J. Hickey, 1 Alyssa Haynes, 1James M. Kikkawa,2 and So-Jung Park1,* 1
Department of Chemistry, University of Pennsylvania, 231 S South 34th Street, Philadelphia, PA 19104 (USA) 2
Department of Physics, University of Pennsylvania, Philadelphia, PA 19104 (USA)
*Corresponding author: [email protected]
Experimental Procedures Synthesis of Block Copolymers. Block copolymers of PAAm-b-PSn were synthesized by the reversible addition fragmentation chain transfer (RAFT) polymerization following a previously reported procedure.1 The molecular weight of synthesized polymers and the degree of polymerization (m, n) were determined by gel permeation chromatography (GPC, Shimadzu). The GPC was equipped with Polymer Laboratories columns (guard, 106, 104, and 5 × 102Å), a S1
UV detector (SPD-10AV), and a refractive index detector (RID-10A) calibrated against linear polystyrene standards in THF. Synthesis of Magnetic Nanoparticles. Iron oxide nanoparticles were synthesized using oleic acid as the stabilizing agent following a modified literature method.2 This procedure was shown to produce a mixture of two different phases, Fe3O4 (magnetite) and γ-Fe2O3 (meghemite).2 Typically, 1.5 g of iron chloride (FeCl3·6H2O, 5.5 mmol, Aldrich, 97%) and 5.2 g of sodium oleate (17 mmol, TCI, 95%) were first added in a 100 mL flask. Subsequently, 20 mL of hexane, 11.5 mL of ethanol, and 8.8 mL of distilled water were added to the flask. The two phase mixture was heated to reflux (~70 oC) and kept at that temperature for four hours, which produced iron-oleate in the organic layer. The upper organic layer was washed three times with 30 mL of water and separated by centrifugation (8,000 rpm, 10 min). After washing, the hexane was evaporated from the dark brown organic layer by rotor evaporation and kept under vacuum overnight (~ 12 hours). The synthesis of 5.6 nm iron oxide nanoparticles was carried out by reacting 5.5 g of iron-oleate and 1.5 g of oleic acid (5.3 mmol, Aldrich, 90%) in 31 g of 1octadecene (Aldrich, 90%) in a 100 mL round-bottom flack. For 14.9 nm particles, a 250 mL round-bottom flask was used. The reaction mixture was heated to 320 oC at a rate of 200 o
C/hour, and kept at that temperature for 30 minutes. The color of the solution turned from dark
brown to black upon the formation of nanoparticles. The resulting solution was cooled to room temperature and nanoparticles were precipitated by adding ethanol (35 mL). The precipitated nanoparticles were collected by centrifuging at 8,000 rpm for 10 min and then redispersed in hexane (10 mL).
The nanoparticles were further purified by adding acetone (35 mL),
centrifuging at 8,000 rpm for 10 min, and redispersing precipitated nanoparticles in hexane (10 mL). The acetone wash was repeated two more times. After the final washing step, the S2
nanoparticles were dissolved in chloroform (10 mL) and centrifuged at low speed (3,000 rpm, 5 min) to remove nanoparticle aggregates.
Finally, chloroform was evaporated off and
nanoparticles were weighed and redissolved in THF (5 mg/mL). The thermogravimetric analysis (TGA) revealed that the weight percent of the iron oxide core in the dried sample was 40 %; the remaining 60 % was oleic acid. After the final purification, the yield of nanoparticle synthesis was determined to be 38 %. Self-assembly of 5.6 nm particles and PAA-b-PS. For magneto-core shell assemblies, 5.6 nm magnetic particles and PAA38-b-PS154 were self-assembled by the slow addition of water to the mixture of nanoparticles and polymers dispersed in a DMF/THF mixture (96.8 vol% DMF: 3.2 vol% THF). In typical experiments, the THF solution of 5.6 nm particles (50 µL of a 2.0 mg/mL) was mixed with a DMF solution of PAA38-b-PS154 (500 µL, 1.1 mg/mL) for the nanoparticle mass percent of 15.9 %.
Then, the total volume of the nanoparticle/polymer
mixture was adjusted to a constant volume of 1.55 mL by adding additional DMF. While stirring, water (600 µL) was slowly added (10 µL per 30 s) to the mixture of nanoparticles and block copolymers. The mixture was kept under stirring for 15 hours before adding additional water (1.5 mL) over 15 minutes. Then, the sample was dialyzed against water for 24 hours, and concentrated by centrifugation (14,000 rpm, 1 hour). After centrifugation, the assemblies were redispersed in 200 µL of deionized water. At the nanoparticle mass percents used in this study, the yields of nanoparticle-encapsulated polymer assemblies were high (close to 100 %) without noticeable precipitation of macroscopic aggregates of nanoparticles and polymers.
For all
experiments, nanoparticle mass percent was adjusted by changing the amount of nanoparticles while keeping the polymer concentration constant. The mass percent of nanoparticles is defined
S3
by 100 times the mass of the dried nanoparticle sample (i.e., iron oxide core and surfactants) over the combined mass of dried nanoparticle sample and polymers. The sample procedure was used for magneto-micelles and magneto-polymersomes formed with THF and dioxane/THF (96.8% dioxane) except THF or dioxane was used in place of DMF. Again, the same procedure was followed to self-assemble nanoparticles with different length polymers (PAA38-b-PS189, PAA38-b-PS73, PAA38-b-PS247). For all experiments, THF (Fisher) used in the self-assembly was purified by distillation, and DMF (Fisher) and dioxane (1,4-Dioxane, 99.8%, extra dry, AcroSeal, Acros Organics) were used without further purification. Co-assembly of two different sized nanoparticles (5.6 nm and 14.9 nm) and PAA38b-PS247. To prepare magneto-core shell assemblies containing two different sized nanoparticles, PAA38-b-PS247, 5.6 nm particles, and 14.9 nm particles were self-assembled by the slow addition of water to the mixture of the three components dispersed in DMF/THF (96.8% DMF). In typical experiments, THF solution of 5.6 nm nanoparticle (25 µL of a 1.0 mg/mL) was mixed with a THF solution of 14.9 nm nanoparticle (25 µL of a 1.0 mg/mL) and mixed with a DMF solution of PAA38-b-PS247 (500.0 µL, 0.31 mg/mL) at a nanoparticle mass percent of 24.4 %. Then, the total volume of the solution was adjusted to 1.55 mL by adding additional DMF. Next, the three components were self-assembled by following the procedure described above. This procedure formed magneto-core shell assemblies containing both 5.6 nm and 14.9 nm particles. On the contrary, when each nanoparticle solution was mixed with polymers before combining the two solutions, nanoparticles stayed separate into different assemblies. For this experiment, a THF solution of 5.6 nm nanoparticle (25 µL of a 1.0 mg/mL) was initially mixed with a DMF solution of PAA38-b-PS247 (250 µL, 0.31 mg/mL). In a separate vial, a THF solution of 14.9 nm S4
nanoparticle (25 µL of a 1.0 mg/mL) was initially mixed with a DMF solution of PAA38-b-PS247 (250 µL, 0.31 mg/mL). The two solutions were then mixed together and the total volume of the nanoparticle/polymer mixture was adjusted to a constant volume (1.55 mL) by adding additional DMF. The same experimental procedures were used to self-assemble 5.6 nm and 14.9 nm nanoparticles with PAA38-b-PS247 in THF only that THF was used in place of DMF. Self-assembly of 5.6 nm particles and PAA38-b-PS73 in dioxane/THF (96.8% dioxane). To generate magneto-polymersomes in high yields, 5.6 nm nanoparticles and PAA38b-PS73 were self-assembled in dioxane/THF (96.8% dioxane) following the same procedure described above for PAA38-b-PS154.
In typical experiments, a THF solution of 5.6 nm
nanoparticles (50 µL of a 1.0 mg/mL) was mixed with a dioxane solution of PAA38-b-PS73 (500
µL, 5.0 mg/mL) for a nanoparticle mass percent of 35.8 %. Then, the total volume of the nanoparticle/polymer mixture was adjusted to 1.55 mL by adding additional dioxane. The polymers and nanoparticles were then self-assembled as stated above. Magnetic relaxivity measurements. The T2 relaxivity times were measured at a series of different sample concentrations using a Bruker mq60 MR relaxometer operating at 1.41 T (60 MHz). The Fe concentration was determined using inductively coupled plasma–atomic emission spectroscopy (ICP-AES) (Spectro Genesis).
Typically, a concentrated stock solution of
magnetic nanocomposites (200 µL of as-prepared assemblies) was added to scintillation vials. The samples in the scintillation vials were then heated at 600 oC for two hours to burn off all organic material. Then, 1 mL of concentrated hydrochloric acid was added to the vials to dissolve all iron oxide nanoparticles. Finally, 9 mL of deionized water was added to the vials. The concentrations of the prepared solutions were then measured using ICP-AES. The T2 relaxivity times were plotted as a function of iron concentration to obtain R2. S5
Blocking Temperature Measurements. The blocking temperature was measured for three samples, magneto-polymersomes, dense iron oxide nanoparticles, and dilute iron oxide nanoparticles. The magneto-polymersome sample was prepared by drying several droplets (30 µL) of aqueous solution of magneto-polymersomes (4mg/mL) on a glass cover slip. The same procedure was repeated for five times to deposit sufficient material. The sample of dense iron oxide nanoparticles was prepared by placing 6 µL of iron oxide nanoparticle solution (10 mg/mL, Chloroform), which contained oleic acid stabilized iron oxide nanoparticles and excess amount of oleic acids, on a glass cover slip and evaporating the solvent to dryness. Fe content of the dense nanoparticle sample was determined using TGA. For the sample of dilute iron oxide nanoparticles, 0.5 g of paraffin was first dispersed in 1 mL of chloroform by sonication and the solution was heated at 70 oC until it became clear (~ 30 sec). Then, a chloroform solution of nanoparticles (5 mg/mL, 80 µL) was added to the paraffin solution. The chloroform was then evaporated to dryness, and the dried sample was scraped out of the vial with non-magnetic tweezers. Magnetic moment measurements were performed using a Quantum Design model MPMS-XL 7 Tesla superconducting quantum interference device (SQUID). To determine the blocking temperature, zero-field cooled measurements were performed for the aforementioned samples.
To account for trapped flux, the magnetic field was zeroed by nulling the
superparamagnetic response to less than 10-5 of the saturation magnetization at a temperature (300 K) well above the blocking temperature (~40 K). Samples were then cooled from 300 K to 10 K, the field was then increased by 10 Oe or 100 Oe, and the magnetic moment was recorded while warming at a rate of 1 K/minute from 10 K to 300 K. The blocking temperature (TB) was identified as the temperature at which the largest moment was produced. We also obtained S6
hysteresis loops at 10 K to determine the saturation magnetic moment after subtraction of the diamagnetic response. Instrumentation. The TEM images were taken using a JEOL 1400 electron microscope, and STEM images were acquired using a JEOL 2010F electron microscope. All EDS mapping was carried out using the JEOL 2010F. Dynamic light scattering (DLS) measurements were taken with a Malvern Zetasizer Nano Series. Power XRD patters were collected using a Rigaku GiegerFlex D/Max-B diffractometer with a Cu anode.
Additional Characterization Data
Figure S1. Transmission electron microscopy (TEM) images of iron oxide nanoparticles. (a) 5.6 ± 0.5 nm particles. (b) 14.9 ± 0.9 nm particles.
S7
Figure S2. Fe intensity line scan obtained by the energy dispersive x-ray spectroscopy for magneto-core shell assemblies (a-d) and magneto-polymersomes (e-h) formed with 5.6 nm particles and PAA38-b-PS189 (0.04 wt %) at the nanoparticle mass percent of 27.1 %. (a) A bright-field STEM image of magneto-core shell assembly. (b) Fe intensity line scan of the magneto-core shell assembly. The two peaks at the left side (c) and the right side (d) of the line scan were fitted with a Gaussian function. The full width at half maximum (FWHM) was determined to be 17.0 ± 4.8 nm from the fitting. (e) A bright-field STEM image of a mixture of magneto-polymersomes. (f) Fe intensity line scan across three magneto-polymersomes. The two peaks at the left side (g) and the right side (h) of the line scan were fitted with a Gaussian function. The FWHM was determined to be 59.8 ± 9.3 nm from the fitting.
Figure S3. DLS data for the assemblies shown in Figure 2 plotted for (a) intensity distribution and (b) number distribution. S8
Table S1. Physical parameters for solvents and polymers. The solubility parameters of PS, acrylic acid, and water are 16.6-20.2, 24.6, and 80.1, respectively.3
Solubility parameter (δ) ([MPa]1/2) Dielectric constant (ε)
DMF
THF
Dioxane
24.8
18.6
20.5
38.2
8.5
2.2
Micelle
Micelle
Vesicle
Polymer geometry*
Polymer morphology without nanoparticles*
*They are experimentally observed morphologies for PAA38-b-PS154 formed at a polymer concentration of 0.1 wt % in three different solvents, 96.8 % DMF containing a small amount of THF, 100% THF, and dioxane containing a small amount of THF. The small amount of THF was added to DMF and dioxane to match the experimental conditions with the assemblies formed with nanoparticles, which are presented in Figure 2. The small amount of THF (3.2 vol%) did not affect the morphologies.
S9
Figure S4. A powder X-ray diffraction (XRD) pattern of 5.6 nm iron oxide particles (black). Known XRD patterns of magnetite (red, JCPDS reference code 00-001-1111) and maghemite (blue, JCPDS reference code 00-004-0755) are plotted as bar graphs along with the measured data for comparison. The measured peak positions of 5.6 nm iron oxide particles match well with the patterns of both phases. It is difficult to differentiate the two phases from the XRD data because the XRD patterns of the two phases are similar and the measured XRD peaks are broad due to the small size of the nanoparticles. The synthetic method used in our study is shown to produce the mixture of magnetite (Fe3O4) and maghemite (γ-Fe2O3).2
S10
Figure S5. Electron diffraction patterns for 5.6 nm iron oxide particles (a) and magnetopolymersomes (b). The lines were added to show that the peak positions remain the same upon the incorporation of particles into the polymersomes. The red, yellow, green, blue, and grey lines represent the 2θ positions of 30.1o (hkl: 220), 35.5o (hkl: 311), 43.0o (hkl: 400), 57.2o (hkl: 511), and 62.7o (hkl: 440), respectively. The diffraction patterns were obtained using selected area electron diffraction (SAED) with the JEOL 2010F TEM at a camera length of 60 cm.
Magnetic Properties To examine whether the close proximity of nanoparticles within magneto-polymersomes can modify the global magnetic behavior, we have compared the zero-field cooled (ZFC) properties of 1) magneto-polymersomes, 2) dense iron oxide nanoparticles, and 3) dilute iron oxide nanoparticles using SQUID magnetometry (Figure S6). The blocking temperature TB is obtained from the peak of the ZFC curve, and is related to particle volume according to kBTB = KV/ln(τ/το), where K is the anisotropy constant, V is the magnetic nanoparticle volume, τ is the measurement time, and το is a characteristic attempt frequency that is difficult to measure but is S11
often estimated so that ln(τ/το) = 25. The anisotropy constant K is observed to depend on particle size.2 However, even when K is assumed to be constant, it is apparent that TB is a sensitive characteristic of particle size. Dilute nanoparticle sample. For the dilute nanoparticle sample where nanoparticles are well-isolated, we find the measured blocking temperature (TB = 40 K) in excellent agreement with similar diameter samples prepared elsewhere by the same method.2 Taking ln(τ/το) = 25, K is calculated to be 1.5 x 106 erg/cm3, which is also in close agreement with prior work, which showed that K increases an order of magnitude beyond its bulk value as diameter drops from 22 to 5 nm.2 Dense nanoparticle sample. For randomly-oriented and distributed particles, TB has been demonstrated both experimentally4 and theoretically5 to increase with the dipolar interaction strength. The dense nanoparticle sample was prepared in the attempt to generate a sample where the dipolar interaction strength is significant.
This sample had the added
advantage that the nanoparticle mass could be easily determined by TGA, giving the particle moment. For this dense nanoparticle sample, the volume fraction of iron oxide core and the particle magnetic moment, m were determined to be 0.10 and 1.5 x 10-17 emu, respectively. Using numerical simulations on 103 magnetic nanoparticles, the average magnitude of the random dipolar field, Bdip, was found to be 90 Oe. This allows us to estimate an upper bound for the fractional shift in TB caused by Bdip as mBdip/KV~10-3, where KV = 1.4 x 10-17 erg. Thus, even for this most dense sample the expected change in TB (~0.1%) is too small to be resolved. As predicted from the calculation, no significant change of TB was observed comparing the dense nanoparticle sample to the dilute nanoparticle sample (Figure S6). S12
Magneto-polymersomes. The volume fraction of iron oxide core in magnetopolymersomes was predicted to be within the range of 0.08 to 0.04 and is lower than that of the dense nanoparticle sample described above (0.10). Thus, the TB of magneto-polymersomes is not expected to significantly change from those of the two nanoparticle samples. Indeed, the data presented in Figure S6 reveal that all three samples exhibited similar blocking temperatures as expected from the calculation. We note that the data for the magneto-polymersomes are shown for 100 Oe, rather than 10 Oe, because the applied field of 100 Oe provided adequate signal-to-noise. Nonetheless, the discussion above showed that the change in the applied field would not cause a significant change in the measured TB. These results indicate that a higher nanoparticle volume fraction is required to induce the dipole-dipole coupling for 5.6 nm particles. In our future studies, assemblies with higher nanoparticle volume fraction or bigger nanoparticles will be examined to investigate the effect of nanoparticle density on the global magnetic behavior of composite materials.
S13
Figure S6. ZFC warming curves for a dilute nanoparticle sample in paraffin (solid) and a dense nanoparticle sample (short dashed) at an applied field of 10 Oe. Also shown is data for drop-cast magneto-polymersomes measured at an applied field of 100 Oe (long dashed). The repeated cooling experiments indicated that the slight decrease in blocking temperature (TB = 37K) of magneto-polymersomes relative to the nanoparticle samples is likely to be caused by small departures from zero magnetization in the zero-field-cooled state, and the blocking temperature of magnetic-polymersomes does not significantly change from the nanoparticle samples.
References 1.
Sanchez-Gaytan, B. L.; Cui, W. H.; Kim, Y. J.; Mendez-Polanco, M. A.; Duncan, T. V.;
Fryd, M.; Wayland, B. B.; Park, S. J. Angew. Chem. Int. Ed. 2007, 46, 9235-9238. 2.
Park, J.; An, K. J.; Hwang, Y. S.; Park, J. G.; Noh, H. J.; Kim, J. Y.; Park, J. H.; Hwang,
N. M.; Hyeon, T. Nature Mater. 2004, 3, 891-895. S14
3.
Brandrup, J. I., Edmund H.; Grulke, Eric A.; Abe, Akihiro; Bloch, Daniel R., Polymer
Handbook. 4th ed. ed.; John Wiley & Sons: New York, 1998. 4.
Luo, W.; Nagel, S. R.; Rosenbaum, T. F.; Rosensweig, R. E. Phys. Rev. Lett. 1991, 67,
2721. 5.
García-Otero, J.; Porto, M.; Rivas, J.; Bunde, A. Phys. Rev. Lett. 2000, 84, 167.
S15
Controlling the Self-Assembly Structure of Magnetic Nanoparticles and Amphiphilic Blockcopolymers: From Micelles to Vesicles Robert J. Hickey, 1 Alyssa Haynes, 1James M. Kikkawa,2 and So-Jung Park1,* 1
Department of Chemistry, University of Pennsylvania, 231 S South 34th Street, Philadelphia, PA 19104 (USA) 2
Department of Physics, University of Pennsylvania, Philadelphia, PA 19104 (USA)
*Corresponding author: [email protected]
Experimental Procedures Synthesis of Block Copolymers. Block copolymers of PAAm-b-PSn were synthesized by the reversible addition fragmentation chain transfer (RAFT) polymerization following a previously reported procedure.1 The molecular weight of synthesized polymers and the degree of polymerization (m, n) were determined by gel permeation chromatography (GPC, Shimadzu). The GPC was equipped with Polymer Laboratories columns (guard, 106, 104, and 5 × 102Å), a S1
UV detector (SPD-10AV), and a refractive index detector (RID-10A) calibrated against linear polystyrene standards in THF. Synthesis of Magnetic Nanoparticles. Iron oxide nanoparticles were synthesized using oleic acid as the stabilizing agent following a modified literature method.2 This procedure was shown to produce a mixture of two different phases, Fe3O4 (magnetite) and γ-Fe2O3 (meghemite).2 Typically, 1.5 g of iron chloride (FeCl3·6H2O, 5.5 mmol, Aldrich, 97%) and 5.2 g of sodium oleate (17 mmol, TCI, 95%) were first added in a 100 mL flask. Subsequently, 20 mL of hexane, 11.5 mL of ethanol, and 8.8 mL of distilled water were added to the flask. The two phase mixture was heated to reflux (~70 oC) and kept at that temperature for four hours, which produced iron-oleate in the organic layer. The upper organic layer was washed three times with 30 mL of water and separated by centrifugation (8,000 rpm, 10 min). After washing, the hexane was evaporated from the dark brown organic layer by rotor evaporation and kept under vacuum overnight (~ 12 hours). The synthesis of 5.6 nm iron oxide nanoparticles was carried out by reacting 5.5 g of iron-oleate and 1.5 g of oleic acid (5.3 mmol, Aldrich, 90%) in 31 g of 1octadecene (Aldrich, 90%) in a 100 mL round-bottom flack. For 14.9 nm particles, a 250 mL round-bottom flask was used. The reaction mixture was heated to 320 oC at a rate of 200 o
C/hour, and kept at that temperature for 30 minutes. The color of the solution turned from dark
brown to black upon the formation of nanoparticles. The resulting solution was cooled to room temperature and nanoparticles were precipitated by adding ethanol (35 mL). The precipitated nanoparticles were collected by centrifuging at 8,000 rpm for 10 min and then redispersed in hexane (10 mL).
The nanoparticles were further purified by adding acetone (35 mL),
centrifuging at 8,000 rpm for 10 min, and redispersing precipitated nanoparticles in hexane (10 mL). The acetone wash was repeated two more times. After the final washing step, the S2
nanoparticles were dissolved in chloroform (10 mL) and centrifuged at low speed (3,000 rpm, 5 min) to remove nanoparticle aggregates.
Finally, chloroform was evaporated off and
nanoparticles were weighed and redissolved in THF (5 mg/mL). The thermogravimetric analysis (TGA) revealed that the weight percent of the iron oxide core in the dried sample was 40 %; the remaining 60 % was oleic acid. After the final purification, the yield of nanoparticle synthesis was determined to be 38 %. Self-assembly of 5.6 nm particles and PAA-b-PS. For magneto-core shell assemblies, 5.6 nm magnetic particles and PAA38-b-PS154 were self-assembled by the slow addition of water to the mixture of nanoparticles and polymers dispersed in a DMF/THF mixture (96.8 vol% DMF: 3.2 vol% THF). In typical experiments, the THF solution of 5.6 nm particles (50 µL of a 2.0 mg/mL) was mixed with a DMF solution of PAA38-b-PS154 (500 µL, 1.1 mg/mL) for the nanoparticle mass percent of 15.9 %.
Then, the total volume of the nanoparticle/polymer
mixture was adjusted to a constant volume of 1.55 mL by adding additional DMF. While stirring, water (600 µL) was slowly added (10 µL per 30 s) to the mixture of nanoparticles and block copolymers. The mixture was kept under stirring for 15 hours before adding additional water (1.5 mL) over 15 minutes. Then, the sample was dialyzed against water for 24 hours, and concentrated by centrifugation (14,000 rpm, 1 hour). After centrifugation, the assemblies were redispersed in 200 µL of deionized water. At the nanoparticle mass percents used in this study, the yields of nanoparticle-encapsulated polymer assemblies were high (close to 100 %) without noticeable precipitation of macroscopic aggregates of nanoparticles and polymers.
For all
experiments, nanoparticle mass percent was adjusted by changing the amount of nanoparticles while keeping the polymer concentration constant. The mass percent of nanoparticles is defined
S3
by 100 times the mass of the dried nanoparticle sample (i.e., iron oxide core and surfactants) over the combined mass of dried nanoparticle sample and polymers. The sample procedure was used for magneto-micelles and magneto-polymersomes formed with THF and dioxane/THF (96.8% dioxane) except THF or dioxane was used in place of DMF. Again, the same procedure was followed to self-assemble nanoparticles with different length polymers (PAA38-b-PS189, PAA38-b-PS73, PAA38-b-PS247). For all experiments, THF (Fisher) used in the self-assembly was purified by distillation, and DMF (Fisher) and dioxane (1,4-Dioxane, 99.8%, extra dry, AcroSeal, Acros Organics) were used without further purification. Co-assembly of two different sized nanoparticles (5.6 nm and 14.9 nm) and PAA38b-PS247. To prepare magneto-core shell assemblies containing two different sized nanoparticles, PAA38-b-PS247, 5.6 nm particles, and 14.9 nm particles were self-assembled by the slow addition of water to the mixture of the three components dispersed in DMF/THF (96.8% DMF). In typical experiments, THF solution of 5.6 nm nanoparticle (25 µL of a 1.0 mg/mL) was mixed with a THF solution of 14.9 nm nanoparticle (25 µL of a 1.0 mg/mL) and mixed with a DMF solution of PAA38-b-PS247 (500.0 µL, 0.31 mg/mL) at a nanoparticle mass percent of 24.4 %. Then, the total volume of the solution was adjusted to 1.55 mL by adding additional DMF. Next, the three components were self-assembled by following the procedure described above. This procedure formed magneto-core shell assemblies containing both 5.6 nm and 14.9 nm particles. On the contrary, when each nanoparticle solution was mixed with polymers before combining the two solutions, nanoparticles stayed separate into different assemblies. For this experiment, a THF solution of 5.6 nm nanoparticle (25 µL of a 1.0 mg/mL) was initially mixed with a DMF solution of PAA38-b-PS247 (250 µL, 0.31 mg/mL). In a separate vial, a THF solution of 14.9 nm S4
nanoparticle (25 µL of a 1.0 mg/mL) was initially mixed with a DMF solution of PAA38-b-PS247 (250 µL, 0.31 mg/mL). The two solutions were then mixed together and the total volume of the nanoparticle/polymer mixture was adjusted to a constant volume (1.55 mL) by adding additional DMF. The same experimental procedures were used to self-assemble 5.6 nm and 14.9 nm nanoparticles with PAA38-b-PS247 in THF only that THF was used in place of DMF. Self-assembly of 5.6 nm particles and PAA38-b-PS73 in dioxane/THF (96.8% dioxane). To generate magneto-polymersomes in high yields, 5.6 nm nanoparticles and PAA38b-PS73 were self-assembled in dioxane/THF (96.8% dioxane) following the same procedure described above for PAA38-b-PS154.
In typical experiments, a THF solution of 5.6 nm
nanoparticles (50 µL of a 1.0 mg/mL) was mixed with a dioxane solution of PAA38-b-PS73 (500
µL, 5.0 mg/mL) for a nanoparticle mass percent of 35.8 %. Then, the total volume of the nanoparticle/polymer mixture was adjusted to 1.55 mL by adding additional dioxane. The polymers and nanoparticles were then self-assembled as stated above. Magnetic relaxivity measurements. The T2 relaxivity times were measured at a series of different sample concentrations using a Bruker mq60 MR relaxometer operating at 1.41 T (60 MHz). The Fe concentration was determined using inductively coupled plasma–atomic emission spectroscopy (ICP-AES) (Spectro Genesis).
Typically, a concentrated stock solution of
magnetic nanocomposites (200 µL of as-prepared assemblies) was added to scintillation vials. The samples in the scintillation vials were then heated at 600 oC for two hours to burn off all organic material. Then, 1 mL of concentrated hydrochloric acid was added to the vials to dissolve all iron oxide nanoparticles. Finally, 9 mL of deionized water was added to the vials. The concentrations of the prepared solutions were then measured using ICP-AES. The T2 relaxivity times were plotted as a function of iron concentration to obtain R2. S5
Blocking Temperature Measurements. The blocking temperature was measured for three samples, magneto-polymersomes, dense iron oxide nanoparticles, and dilute iron oxide nanoparticles. The magneto-polymersome sample was prepared by drying several droplets (30 µL) of aqueous solution of magneto-polymersomes (4mg/mL) on a glass cover slip. The same procedure was repeated for five times to deposit sufficient material. The sample of dense iron oxide nanoparticles was prepared by placing 6 µL of iron oxide nanoparticle solution (10 mg/mL, Chloroform), which contained oleic acid stabilized iron oxide nanoparticles and excess amount of oleic acids, on a glass cover slip and evaporating the solvent to dryness. Fe content of the dense nanoparticle sample was determined using TGA. For the sample of dilute iron oxide nanoparticles, 0.5 g of paraffin was first dispersed in 1 mL of chloroform by sonication and the solution was heated at 70 oC until it became clear (~ 30 sec). Then, a chloroform solution of nanoparticles (5 mg/mL, 80 µL) was added to the paraffin solution. The chloroform was then evaporated to dryness, and the dried sample was scraped out of the vial with non-magnetic tweezers. Magnetic moment measurements were performed using a Quantum Design model MPMS-XL 7 Tesla superconducting quantum interference device (SQUID). To determine the blocking temperature, zero-field cooled measurements were performed for the aforementioned samples.
To account for trapped flux, the magnetic field was zeroed by nulling the
superparamagnetic response to less than 10-5 of the saturation magnetization at a temperature (300 K) well above the blocking temperature (~40 K). Samples were then cooled from 300 K to 10 K, the field was then increased by 10 Oe or 100 Oe, and the magnetic moment was recorded while warming at a rate of 1 K/minute from 10 K to 300 K. The blocking temperature (TB) was identified as the temperature at which the largest moment was produced. We also obtained S6
hysteresis loops at 10 K to determine the saturation magnetic moment after subtraction of the diamagnetic response. Instrumentation. The TEM images were taken using a JEOL 1400 electron microscope, and STEM images were acquired using a JEOL 2010F electron microscope. All EDS mapping was carried out using the JEOL 2010F. Dynamic light scattering (DLS) measurements were taken with a Malvern Zetasizer Nano Series. Power XRD patters were collected using a Rigaku GiegerFlex D/Max-B diffractometer with a Cu anode.
Additional Characterization Data
Figure S1. Transmission electron microscopy (TEM) images of iron oxide nanoparticles. (a) 5.6 ± 0.5 nm particles. (b) 14.9 ± 0.9 nm particles.
S7
Figure S2. Fe intensity line scan obtained by the energy dispersive x-ray spectroscopy for magneto-core shell assemblies (a-d) and magneto-polymersomes (e-h) formed with 5.6 nm particles and PAA38-b-PS189 (0.04 wt %) at the nanoparticle mass percent of 27.1 %. (a) A bright-field STEM image of magneto-core shell assembly. (b) Fe intensity line scan of the magneto-core shell assembly. The two peaks at the left side (c) and the right side (d) of the line scan were fitted with a Gaussian function. The full width at half maximum (FWHM) was determined to be 17.0 ± 4.8 nm from the fitting. (e) A bright-field STEM image of a mixture of magneto-polymersomes. (f) Fe intensity line scan across three magneto-polymersomes. The two peaks at the left side (g) and the right side (h) of the line scan were fitted with a Gaussian function. The FWHM was determined to be 59.8 ± 9.3 nm from the fitting.
Figure S3. DLS data for the assemblies shown in Figure 2 plotted for (a) intensity distribution and (b) number distribution. S8
Table S1. Physical parameters for solvents and polymers. The solubility parameters of PS, acrylic acid, and water are 16.6-20.2, 24.6, and 80.1, respectively.3
Solubility parameter (δ) ([MPa]1/2) Dielectric constant (ε)
DMF
THF
Dioxane
24.8
18.6
20.5
38.2
8.5
2.2
Micelle
Micelle
Vesicle
Polymer geometry*
Polymer morphology without nanoparticles*
*They are experimentally observed morphologies for PAA38-b-PS154 formed at a polymer concentration of 0.1 wt % in three different solvents, 96.8 % DMF containing a small amount of THF, 100% THF, and dioxane containing a small amount of THF. The small amount of THF was added to DMF and dioxane to match the experimental conditions with the assemblies formed with nanoparticles, which are presented in Figure 2. The small amount of THF (3.2 vol%) did not affect the morphologies.
S9
Figure S4. A powder X-ray diffraction (XRD) pattern of 5.6 nm iron oxide particles (black). Known XRD patterns of magnetite (red, JCPDS reference code 00-001-1111) and maghemite (blue, JCPDS reference code 00-004-0755) are plotted as bar graphs along with the measured data for comparison. The measured peak positions of 5.6 nm iron oxide particles match well with the patterns of both phases. It is difficult to differentiate the two phases from the XRD data because the XRD patterns of the two phases are similar and the measured XRD peaks are broad due to the small size of the nanoparticles. The synthetic method used in our study is shown to produce the mixture of magnetite (Fe3O4) and maghemite (γ-Fe2O3).2
S10
Figure S5. Electron diffraction patterns for 5.6 nm iron oxide particles (a) and magnetopolymersomes (b). The lines were added to show that the peak positions remain the same upon the incorporation of particles into the polymersomes. The red, yellow, green, blue, and grey lines represent the 2θ positions of 30.1o (hkl: 220), 35.5o (hkl: 311), 43.0o (hkl: 400), 57.2o (hkl: 511), and 62.7o (hkl: 440), respectively. The diffraction patterns were obtained using selected area electron diffraction (SAED) with the JEOL 2010F TEM at a camera length of 60 cm.
Magnetic Properties To examine whether the close proximity of nanoparticles within magneto-polymersomes can modify the global magnetic behavior, we have compared the zero-field cooled (ZFC) properties of 1) magneto-polymersomes, 2) dense iron oxide nanoparticles, and 3) dilute iron oxide nanoparticles using SQUID magnetometry (Figure S6). The blocking temperature TB is obtained from the peak of the ZFC curve, and is related to particle volume according to kBTB = KV/ln(τ/το), where K is the anisotropy constant, V is the magnetic nanoparticle volume, τ is the measurement time, and το is a characteristic attempt frequency that is difficult to measure but is S11
often estimated so that ln(τ/το) = 25. The anisotropy constant K is observed to depend on particle size.2 However, even when K is assumed to be constant, it is apparent that TB is a sensitive characteristic of particle size. Dilute nanoparticle sample. For the dilute nanoparticle sample where nanoparticles are well-isolated, we find the measured blocking temperature (TB = 40 K) in excellent agreement with similar diameter samples prepared elsewhere by the same method.2 Taking ln(τ/το) = 25, K is calculated to be 1.5 x 106 erg/cm3, which is also in close agreement with prior work, which showed that K increases an order of magnitude beyond its bulk value as diameter drops from 22 to 5 nm.2 Dense nanoparticle sample. For randomly-oriented and distributed particles, TB has been demonstrated both experimentally4 and theoretically5 to increase with the dipolar interaction strength. The dense nanoparticle sample was prepared in the attempt to generate a sample where the dipolar interaction strength is significant.
This sample had the added
advantage that the nanoparticle mass could be easily determined by TGA, giving the particle moment. For this dense nanoparticle sample, the volume fraction of iron oxide core and the particle magnetic moment, m were determined to be 0.10 and 1.5 x 10-17 emu, respectively. Using numerical simulations on 103 magnetic nanoparticles, the average magnitude of the random dipolar field, Bdip, was found to be 90 Oe. This allows us to estimate an upper bound for the fractional shift in TB caused by Bdip as mBdip/KV~10-3, where KV = 1.4 x 10-17 erg. Thus, even for this most dense sample the expected change in TB (~0.1%) is too small to be resolved. As predicted from the calculation, no significant change of TB was observed comparing the dense nanoparticle sample to the dilute nanoparticle sample (Figure S6). S12
Magneto-polymersomes. The volume fraction of iron oxide core in magnetopolymersomes was predicted to be within the range of 0.08 to 0.04 and is lower than that of the dense nanoparticle sample described above (0.10). Thus, the TB of magneto-polymersomes is not expected to significantly change from those of the two nanoparticle samples. Indeed, the data presented in Figure S6 reveal that all three samples exhibited similar blocking temperatures as expected from the calculation. We note that the data for the magneto-polymersomes are shown for 100 Oe, rather than 10 Oe, because the applied field of 100 Oe provided adequate signal-to-noise. Nonetheless, the discussion above showed that the change in the applied field would not cause a significant change in the measured TB. These results indicate that a higher nanoparticle volume fraction is required to induce the dipole-dipole coupling for 5.6 nm particles. In our future studies, assemblies with higher nanoparticle volume fraction or bigger nanoparticles will be examined to investigate the effect of nanoparticle density on the global magnetic behavior of composite materials.
S13
Figure S6. ZFC warming curves for a dilute nanoparticle sample in paraffin (solid) and a dense nanoparticle sample (short dashed) at an applied field of 10 Oe. Also shown is data for drop-cast magneto-polymersomes measured at an applied field of 100 Oe (long dashed). The repeated cooling experiments indicated that the slight decrease in blocking temperature (TB = 37K) of magneto-polymersomes relative to the nanoparticle samples is likely to be caused by small departures from zero magnetization in the zero-field-cooled state, and the blocking temperature of magnetic-polymersomes does not significantly change from the nanoparticle samples.
References 1.
Sanchez-Gaytan, B. L.; Cui, W. H.; Kim, Y. J.; Mendez-Polanco, M. A.; Duncan, T. V.;
Fryd, M.; Wayland, B. B.; Park, S. J. Angew. Chem. Int. Ed. 2007, 46, 9235-9238. 2.
Park, J.; An, K. J.; Hwang, Y. S.; Park, J. G.; Noh, H. J.; Kim, J. Y.; Park, J. H.; Hwang,
N. M.; Hyeon, T. Nature Mater. 2004, 3, 891-895. S14
3.
Brandrup, J. I., Edmund H.; Grulke, Eric A.; Abe, Akihiro; Bloch, Daniel R., Polymer
Handbook. 4th ed. ed.; John Wiley & Sons: New York, 1998. 4.
Luo, W.; Nagel, S. R.; Rosenbaum, T. F.; Rosensweig, R. E. Phys. Rev. Lett. 1991, 67,
2721. 5.
García-Otero, J.; Porto, M.; Rivas, J.; Bunde, A. Phys. Rev. Lett. 2000, 84, 167.
S15
