Photochemical Generation of Ag, Pd, and Pt Particles in Octane - The

Article pubs.acs.org/JPCC

Photochemical Generation of Ag, Pd, and Pt Particles in Octane Dan R. Clary,† Mahdi Nabil,‡ Mahmoud M. Sedeh,‡ Yousef El-Hasadi,‡ and G. Mills*,† †

Department of Chemistry and Biochemistry, Auburn University, Auburn, Alabama 36849, United States Department of Mechanical Engineering, Auburn University, Auburn, Alabama 36849, United States

‡

ABSTRACT: Metal crystallites formed upon UV irradiation of benzophenone (BP) solutions in octane containing oleoyl sarcosine as a particle stabilizer and Ag neodecanoate (AgOOR), as well as the complexes Pd(acac)2 and Pt(acac)2, with acac = acetylacetonate anion. Initial quantum yields of metal formation for Pd(acac)2 and Pt(acac)2 were 7 × 10−3 and 4.5 × 10−3, respectively, which are about 7 times higher than those for analogous reactions conducted in the absence of the BP photosensitizer. Direct irradiation of AgOOR, on the other hand, resulted in no reaction, but silver particles were formed with a quantum yield of 1.2 in the presence of BP. Whereas the Pd colloids decay within 1 week of preparation, the Ag and Pt particles have remained stable, thus far, in octane for more than 4 months. The resulting octane colloids were evaluated for enhancements in thermal conductivity using the thermal hot disk method. Enhancements of up to 10% were observed for the Ag and Pt systems at metal concentrations of 5 mM, which are far larger than what Maxwell’s theory predicts for a colloid of low volume fraction (∼5 × 10−5 vol %).

■

INTRODUCTION The use of light as the driving force in metal salt reduction provides an interesting tool for the generation of colloidal particles, which has recently been reviewed.1 For instance, Ag and Pt colloids have been prepared from aqueous solutions of Ag(I) and Pt(IV) ions by direct photolysis.2,3 Alternatively, αhydroxy radicals resulting from illumination of benzophenone or acetone solutions have been employed to reduce metal ions, forming colloids.4 Semiconductors, such as CdSe and CdS, have been employed as photosensitizers in the generation of aqueous colloids of Pd and Pt,5 whereas TiO2 particles have similarly been used in aqueous alcohol solution to produce Ag and Pt particles.6,7 Pt(acac)2 was made water-soluble by inclusion inside the cavity of β-cyclodextrin, and irradiation of the resulting complex with both UV and visible light resulted in metal particles.8 Likewise, direct irradiation of H2PtCl6 and PdCl2 with UV photons resulted in metal ion reduction.9,10 However, scant information is available regarding the photochemical generation of Ag, Pd, or Pt particles in hydrocarbon solvents. The closest available reference involved the photochemical generation of Ag and Pt particles in aqueous ethanol solutions that were later isolated and redispersed in benzene and THF.11 Filtration was required to remove the fraction of particles that initially precipitated. Previous attempts to generate noble metal colloids in nonpolar solvents have been dominated by two-phase systems in which aqueous metal salts are extracted into an organic layer with the aid of phase-transfer agents, such as phosphines or quaternary amines.12 Salt reduction ensues by the addition of a strong reducing agent, such as NaBH4 or hydrazine.13−23 Provided that salt dissolution can be achieved, in situ reduction can, at times, be made possible in the organic phase by the input of thermal energy. Silver lactate, for example, was © 2012 American Chemical Society

thermally decomposed in mineral oil in the presence of Korantin SH as a reductant as well as a particle stabilizer to produce particles in large concentrations that were stable for several weeks.24 Thermal decomposition of various organicsoluble salts of silver and gold have produced isolable particles that could be dispersed in numerous solvents with low dielectric constants, including hexane, heptane, toluene, and others.25−27 Reduction of Pd(acetate)2 and Pd(acac)2 was also achieved in boiling aliphatic and aromatic alcohols.28 Other synthetic routes involving nonpolar media include the electrochemical generation of Ag particles in dimethyl sulfoxide,29 the formation of silver-encapsulated reverse micelles that could be isolated, then dispersed in hexane,30 and the reduction of silver acetate by phenyl hydrazine in toluene.31 Heat-generating processes and machines, such as materials production, electronic devices, combustion engines, fuel cells, batteries, and power plants, require a means by which excess heat can be removed. Traditional cooling processes involve exposing the system to a medium capable of extracting and transferring heat to the surroundings. Such a medium may include forced air, water, ethylene glycol, or other cooling liquid; however, many of these compounds have a low thermal conductivity (TC) relative to the materials composing the system, such as metals, silicon, and ceramics.32,33 These demands create a need for improved cooling technologies. In 1873, Maxwell proposed dispersing metallic particles in a given solvent to produce a medium with enhanced electrical conductivity,34 the utility of which was later extended to heat conduction.35 This concept of enhanced heat transfer was Received: October 19, 2011 Revised: February 28, 2012 Published: April 5, 2012 9243

dx.doi.org/10.1021/jp210051j | J. Phys. Chem. C 2012, 116, 9243−9250

The Journal of Physical Chemistry C

Article

inside an inert atmosphere to avoid contamination by water and air. Absorbance spectra were acquired on a Shimadzu UV-2450 spectrophotometer; electron micrographs were obtained on a Zeiss EM10 transmission electron microscope at 60 kV. Selected area electron diffraction data resulted from analysis on a Hitachi HF-2000 TEM at 200 kV. Thermal conductivity measurements were conducted with a Hot Disk Transient Plane Source TC detector.43 Aliquots (10 mL) of each colloid were deposited in Pyrex sample holders, fitted with a cover containing a slit to accommodate the Thermal Hot Disk sensor, and the containers were then inserted into matching cavities within an aluminum block submerged in a thermostatic bath. Samples were equilibrated until they remained at the desired temperature, ± 0.1 °C, for at least 5 min, and the reported TC value is the average of five independent measurements, the standard deviation of which was used to approximate the experimental error. Specimens were prepared for TEM observation by dip-coating a 300 mesh Cu/Formvar grid into a 0.1−0.5 mM colloid (either made directly or by dilution of more concentrated colloids) and oven drying at 60 °C under vacuum for 10 min. Photochemical reactions were conducted in a 1 cm path length cuvette fitted with a quartz-to-pyrex graded seal. Solutions were prepared in the optical cell within a nitrogen-filled glovebag and sealed with a rubber septum. Sample irradiation was achieved by centering the cell inside of a Rayonet 100 circular illuminator equipped with 16 RPR-3500A lamps generating photons of λ = 350 ± 15 nm, and chemical actinometry was conducted using Aberchrome 540.44

founded on the knowledge that metals tend to have much higher thermal conductivities than common liquids. Copper, for example, has a conductivity about 700 times larger than that of water and about 3000 times greater than engine oil.36 Novel heat-transfer liquids were initially attempted by dispersing micro- to millimeter-sized particles, but attempts were met with little success due to immediate sedimentation of the particles. Over the past decade, however, researchers have turned to traditional colloid chemistry for inspiration. After suspending nanometer-sized copper(II) oxide in water and measuring the resulting enhancement of TC of the mixture, the term “nanofluid” was coined and has since become widely accepted.37 Strictly speaking, a nanofluid is a colloid whose design is specific to the study and application of heat-transfer technology. An additional strategy for the management of excess heat involves the use of phase change materials (PCMs) that capitalize on the latent heat of a material to extract and store thermal energy. The set point of storage can be easily tailored in hydrocarbon-based PCMs by simply lengthening or shortening the carbon chain. As in nanofluids, the rate of heat transfer through a PCM may be enhanced by the stable dispersion of metallic particles.38 Therefore, PCMs based on hydrocarbons that were enhanced by metal nanoparticles seemed interesting. Such an idea inspired the preparation of metal−octane nanofluids as a preemptive model for future PCMs. A previous study demonstrated the successful photogeneration of Cu particles in octane, which produced highly stable hydrocarbon-based colloids.39 Because of the air sensitivity of the Cu particles, however, efforts were turned toward the more robust noble metals. Investigations of nanofluids have traditionally been limited to highly polar media, such as ethylene glycol and water, with little published work utilizing nonpolar solvents.40 Silver nanoparticles were prepared by thermal reduction of aqueous AgNO3 with ascorbic acid, isolated by centrifugation, and redispersed in kerosene. Enhancements of the thermal conductivity of the resulting nanofluid were evaluated using a transient calorimeter. The tests showed an increase of nearly 20% at 50 °C for a colloid containing 0.5 mass % Ag.41 Reported in the present study, sponsored by an agency of the U.S. Government,42, is a method for the photochemical generation of colloidal Ag, Pd, and Pt in octane and the resulting enhancement of thermal conductivity as measured by the thermal hot disk method.

■

RESULTS AND DISCUSSION Irradiation of octane solutions containing AgOOR, Pd(acac)2, or Pt(acac)2 with 350 nm photons in the presence of BP and OS resulted in metal particles with nanometer dimensions. Absorbance spectra of the resulting yellow (Ag), dilute black (Pd), and light brown (Pt) solutions are shown in Figure 1.

■

EXPERIMENTAL SECTION Pd(acac)2 and Pt(acac)2 were purchased from VWR, silver neodecanoate and oleoyl sarcosine from TCI America, and all other chemicals from Sigma-Aldrich. Rigorous purification of OS proved to be crucial to obtaining reproducible results. To remove water and other volatiles, the sarcosine was twice diluted in excess dry toluene, followed by solvent removal under partial pressure and continued distillation for 4 h. Benzophenone (BP) was recrystallized from methanol and water while the remaining reagents were used as received. Prior to use, octane was scanned from 900 to 200 nm to ensure the absence of light-absorbing impurities. All glassware was treated with fresh piranha acid, rinsed thoroughly with deionized water, and oven-dried. Stock solutions of Pd(acac)2 and Pt(acac)2 in dry toluene were prepared daily in the concentration range of 25−50 mM, whereas working stocks of AgOOR were prepared daily in anhydrous octane. All reagents were stored and handled

Figure 1. Absorbance spectra of metal particles generated by 350 nm illumination of air-free octane solutions containing 0.1 mM AgOOR, 0.5 mM Pd(acac)2, and 0.5 mM Pt(acac)2 in the presence of 10 mM OS.

The sharp plasmon peak at 400 nm for Ag is due to the collective oscillation of electrons about the particle surface, whereas the broad absorption continua throughout the visible region in the Pd and Pt spectra result from the superposition of interband excitations with plasmon resonances.45 Figure 2A−C depicts electron micrographs demonstrating that the particles possess a fairly spherical morphology. Size distributions resulted from counting a minimum of 250 particles. The corresponding histograms are shown in Figure 2D−F and reveal mean particle diameters of 11, 3, and 12.5 nm for Ag, Pd, and Pt, respectively. TEM determinations revealed that the 9244



Article

Figure 2. TEM images of Ag (A), Pd (B), and Pt (C) particles revealing a spherical morphology. Histograms of particle sizes with mean diameters of 11, 3, and 12.5 nm, respectively (D−F). Corresponding electron diffraction patterns verifying particle crystallinity (G−I).

histograms and particle shape remained practically invariant when the colloids were prepared using metal ion concentrations ranging from 0.1 to 5 mM. Selected area electron diffraction (SAED) was employed to verify the crystallinity of the particles. The Debye rings shown in Figure 2G−I are in good agreement with the reflections published by the JCPDS.46 Colloids containing Ag and Pt with metal concentrations of 0.1 and 0.5 mM show no evidence of flocculation after 4 months, according to visual inspection, but the Pd particles precipitate within 1 week of preparation, which is in contrast to the aqueous Pd colloids prepared in our lab.47 The instability of the organic-phase Pd particles is likely a result of the small material dimensions, which results in energetic surfaces that can cause rapid aggregation and subsequent precipitation. Direct irradiation of Pd(acac)2 and Pt(acac)2 in the absence of BP resulted in metal particles, whereas Ag required the aid of a photosensitizer. BP-sensitized metal ion reduction in octane is expected to occur via photogenerated ketyl radicals, as previously documented for the photoreduction of Cu(oleate)2 in hydrocarbon solvents.39 Absorption of a 350 nm photon by BP results in an excited singlet state that undergoes intersystem crossing with near 100% efficiency to yield a triplet excited state of BP (3BP*), as shown in eq 1.48 The excited triplet state is capable of hydrogen abstraction from alcohols, BP + hν → 1BP* → 3BP*

amines, and hydrocarbons, the latter two of which are present in the current system. The first step in 3BP* reduction by tertiary amines seems to involve electron transfer from the nonbonding pair of nitrogen to form a short-lived contact ion pair.49,50 In nonpolar media, ultrafast proton transfer follows to yield the ketyl and alkylaminyl radicals.51 Data from the study on the photoreduction of Cu(oleate)2 in octane indicated that 3 BP* was indeed quenched by OS.39 However, the result of the quenching process was inhibition of the Cu(II) reduction, implying that OS is not an efficient H-atom donor. 3BP* is also known to abstract H atoms directly from alkanes to form the BP ketyl radical (BPH•) and alkyl radicals (RH•).52 In fact, the dominant pathway for the ketyl BP radical formation in octane was shown to involve abstraction of a H atom from the solvent by 3BP*, as illustrated in eq 239 3

BP* + RH 2 → BPH• + HR•

(2)

Since oxidation of octane by 3BP* in CCl4 at room temperature occurs with a rate constant of k = 4.7 × 105 M−1 s−1,52 the same value is assigned to k2. Thus, k2′ = k2[C8H18] = 2.9 × 106 s−1 when the concentration of pure octane (6.15 M) is used, which is a reasonable approximation because the concentrations of all solutes are relatively low. Evaluation of metal formation rate was conducted using the initial rate method, denoted ri. Where appropriate, initial

(1) 9245



Article

quantum yields, φi = ri/(ρIo), are additionally reported, where ρ is the fraction of photons absorbed by the chromophore at the given concentration and Io is the light intensity determined in μM(hν)/s. Illumination of Pd(acac)2 and Pt(acac)2 in the absence of BP is known to induce homolytic cleavage of the ligandmetal bond, according to eqs 3 and 4, resulting in a metal atom and two ligand radicals (acac•).8 These radicals then abstract H atoms from a donor, such as the C8H18 solvent, to yield two Hacac molecules, as shown in eq 5, where R denotes the oxidized hydrogen atom donor, in this case, octane. M(acac)2 + hν → M(acac) + acac•

(3)

M(acac) + hν → M + acac•

(4)

2acac• + RH 2 → 2Hacac + R

(5)

both cases, initial quantum yields were at least 1 order of magnitude larger than values obtained under the same conditions in dichloromethane with λ ≥ 300 nm.53 Analogous experiments performed in the presence of BP and Io = 20 μM/s yielded faster photoreactions by a factor of about 3 with ri = 1.3 × 10−7 M/s and φi = 7 × 10−3 for Pd, whereas ri = 9.4 × 10−8 M/s and φi = 4.5 × 10−3 for Pt. Illustrated in Figure 4A is the

Because of the absence of a sharp absorption band in the spectra of Pd and Pt colloids, initial rates and quantum yields for the formation of the particles were obtained by following the optical density at 450 nm as a function of time. Calculation of the kinetic parameters employed the extinction coefficient per mole of metal atoms (ε450) evaluated for each colloid at this wavelength, ε450 = 318 M−1 cm−1 for Pd and ε450 = 1.2 × 104 M−1 cm−1 for Pt. Figure 3A depicts the spectral changes

Figure 4. Evolution of optical spectra during the photolysis of octane solutions containing 0.5 mM M(acac)2, 2 mM BP, and 10 mM OS, and Io = 53 μM/s, where M = Pd2+ (A) or Pt2+ (B). Presented in the insets are the corresponding initial rate plots.

spectral evolution during the photosensitized formation of Pd; comparison of these optical data with those depicted in Figure 3A indicates a substantial increase in the intensity of the spectrum at all wavelengths resulting in ε450 = 1 × 104 M−1 cm−1. Figure 4B presents the spectral evolution during the photogeneration of colloidal Pt, which exhibited a spectrum analogous to that displayed in Figure 3B with ε450 = 1.4 × 104 M−1 cm−1. Included in the insets of each figure are the corresponding initial rate plots. The BP-sensitized photoreduction of Pd(acac)2 resulted in colloidal solutions with an extinction coefficient 3 times larger than that which was observed in Pd colloids prepared by direct irradiation. To verify that the smaller extinction was not due to incomplete ion reduction, a control experiment was conducted. A solution containing 0.5 mM Pd2+ and 10 mM OS in octane was irradiated until the evolution of optical density at 450 nm was complete. Addition of 4 mM BP and further irradiation for 1 h induced no further increase in absorbance, indicating that all dissolved metal ions had been reduced. The decrease in extinction cross section is expected if smaller particles are generated, which exhibit weaker optical transitions.45 TEM analyses of the corresponding samples confirmed this assumption as they revealed particle sizes ≤ 2 nm; smaller sizes are below the resolution limit of the employed instrument.

Figure 3. Evolution of optical spectra during the direct photolysis of octane solutions containing 0.5 mM M(acac)2 and 10 mM OS, Io = 53 μM/s, where M = Pd2+ (A) or Pt2+ (B). Shown in the insets are the corresponding initial rate plots.

recorded during the formation of metallic Pd upon photolysis without BP. An analogous experiment with Pt(acac)2 yielded the spectral evolution illustrated in Figure 3B. Presented in the insets of both figures are the corresponding rate plots. When dry, air-free octane solutions containing 0.5 mM M(acac)2 and 10 mM OS were irradiated with Io = 55 μM/s, metal formation occurred with an initial rate of ri = 4.7 × 10−8 M/s (φi = 9 × 10−4) for Pd and ri = 3.4 × 10−8 M/s (φi = 6 × 10−4) for Pt. In 9246



Article

The spectral evolution resulting from photolysis with Io = 29 μM/s of an octane solution containing 0.1 mM AgOOR, 0.1 mM BP, and 10 mM OS is shown in Figure 5; the inset depicts

Figure 6. Absorbance spectra corresponding to (A) a mature plasmon band obtained after 1.75 min of irradiation time and under the conditions listed in Figure 5, (B) broadening of the previous signal after a total irradiation time of 6 min with the formation of a signal at 325 nm, and (C) further broadening of the plasmon and UV peaks of (B) after the further addition of 0.1 mM BP and continued irradiation of 10 additional minutes. Spectrum (D) acquired after exposing the sample from the previous spectrum to pure O2 and allowing it to react in the dark for 35 min.

Figure 5. Evolution of optical spectra during the photolysis of air-free octane solutions containing 0.1 mM AgOOR, 0.1 mM BP, and 10 mM OS, and Io = 29 μM/s; the inset shows an initial rate plot.

the corresponding initial rate plot. Formation of Ag crystallites was characterized by a broad initial peak centered at 422 nm, which sharpened while simultaneously shifting to shorter wavelengths during the course of the photoreaction. Similar observations have been made when Ag+ is photoreduced in aqueous solutions sensitized by a polymeric benzophenone.54 The initially broad peak has been ascribed to the coexistence of small metal clusters and particles, while the peak strengthening results from their growth to form nanometer-sized crystallites exhibiting a plasmon band. In the present system, the wavelength of maximum absorption (λmax) was 398 nm and ε398 = 1.2 × 104 M−1 cm−1. Silver formation occurred with an initial rate of 1.2 × 10−6 M/s and φi = 1.2, the latter being about 20 times larger than the value reported for the photoreduction of argentate ions by the polymeric BP sensitizer in water.54 The photochemical synthesis of stable Ag particles was very sensitive to reaction conditions, particularly the rigorous exclusion of air and water. For instance, when a solution was first photolyzed for a time shorter than that needed for complete Ag+ reduction and then illuminated again, erratic results were obtained, including formation of unstable particles that exhibited broad optical signals centered above 500 nm. Hence, reproducible synthesis of stable Ag crystallites was achieved via subjecting an unexposed solution to a single continuous illumination period without interruptions. In fact, each spectrum presented in Figure 5 was acquired using this procedure. Similar detrimental effects were occasionally noticed when colloids were overexposed to light. Formation of the plasmon band shown in Figures 5 and 6, spectrum A, was complete in 1.75 min, but extended irradiation beyond the point of completion induced optical changes and particle precipitation within 72 h. Figure 6, spectrum B, illustrates this behavior where excessive illumination caused the plasmon band to broaden and decrease in intensity, shifting to λmax = 510 nm, as well as the emergence of a signal centered at 325 nm. Subsequent photolysis of the sample induced no spectral changes, but addition of 0.1 mM BP to the solution and irradiation for another 10 min yielded spectrum C shown in Figure 6. As before, the visible absorption broadened and decreased in intensity while the strength of the 325 nm signal increased. A possible origin for the UV absorption are light-absorbing transients (LATs) that exhibit a signal with λmax = 325 nm.55

These species form via dimerization of ketyl radicals, or from combination of ketyl plus alky radicals, and are known to oxidize fast in the presence of oxygen. To test this possibility, the solution that yielded spectrum C of Figure 6 was exposed to pure O2, shaken repeatedly, and then left to react for 35 min. As shown in Figure 6, spectrum D, the signal with λmax = 325 nm remained unchanged after this treatment, eliminating the possibility of a contribution from a LAT. Broadening and shifting of the metal absorption could result from particle agglomeration and growth. However, TEM experiments with colloids possessing either the sharp plasmon or the broad optical signal with λmax = 510 nm revealed no significant changes in size distribution or particle shape. On the other hand, images from colloids exhibiting the broad absorption yielded particle boundaries less well defined than those shown in Figure 2G. A plausible explanation for such a phenomenon is adsorption of some material on the Ag surfaces. Absorptions at short wavelengths have been frequently detected during the formation of Ag colloids in H2O and CH3OH. Radiolytic reduction of Ag+ generated transient ionic clusters consisting of Ag atoms bound to silver ions; these species exhibit UV absorptions in some cases with λmax = 325 nm.56 Poly(acrylic acid), PAA, can act as a stabilizer, enabling formation of long-lived Ag clusters that persisted when O2 was present.57 Photoreduction of Ag+ in blend films of PAA with poly(vinyl alcohol) also generated small Ag clusters stable in the presence of air.58 On the other hand, addition of iodine to aqueous silver colloids resulted in signals at short wavelengths and broadening of the plasmon band similar to the observations made in the present study.59 These findings were interpreted in terms of oxidation of surface metal atoms with formation of a surficial AgI layer. Analogous changes in the particle plasmon were noticed upon exposing aqueous silver colloids to various oxidizing species.60 In view of the earlier observations, the optical changes depicted in Figure 6 may be derived from interactions that originate when small Ag clusters are present on the surface of the silver crystallites. Formation of ionic Ag clusters in a nonpolar solvent is unexpected. The presence of adventitious water could stabilize such species and also explain the somewhat erratic behavior of the silver system. The quantum efficiency of the photosensitized reduction of AgOOR in octane is, at least, 170 times higher than the quantum yields for Pd(acac)2 and Pt(acac)2. Such a large 9247



Article

the range of 10 °C ≤ T ≤ 30 °C, the TC values followed the trend Ag > Pt > Pd, which agrees qualitatively with the differences in thermal conductivities of the bulk metals.66 The efficacy of nanofluids is typically evaluated in terms of the relative thermal conductivity or TC enhancement (ΔTC), where ΔTC = knf/ksol, knf is the TC of the colloid, and ksol is the TC of octane. Presented in Figure 7B is a plot of ΔTC versus temperature for each colloid; in all cases, the relative thermal conductivity increased smoothly with rising T. Maximum enhancements of ∼10% occurred at 50 °C for both Ag and Pt systems and 7% for the Pd colloid, with an estimated experimental error of ±1%. An interesting issue is the substantial difference between the ΔTC data and the values derived from the theory of Maxwell. For colloids composed of [M] = 5 mM, this simple model predicts a small enhancement of about 1.5 × 10−2 %,67 which is approximately 600 times lower than the results of the present study. This divergence is not surprising given that numerous experimental results are at odds with the classical model. Some of the data have been rationalized on the premise that particles dispersed in a liquid medium form linear aggregates that provide a new path for heat conduction.67 However, extensive TEM experiments failed to detect aggregates in the octane colloids. Another noteworthy finding is that the TC enhancements determined in octane are higher than the corresponding values for aqueous Ag, Pt, and Pd colloids of 2, 1, and 2%, respectively.47 The difference in ΔTC for aqueous versus hydrocarbon colloids originates from the TC value of octane being 6 times lower than that of water.64 Hence, the metal particles are able to exert a larger contribution to the heat conduction in the case of octane as compared with aqueous systems.

difference in photochemical efficiency should be related, in principle, to the chemical reactivity of the metal precursors. Redox potentials determined in acetonitrile will be used since relevant data in octane are not available. The redox potential of Ag+ in this solvent is 0.39 V vs NHE,61 whereas the value for Pd(acac)2 is −0.72 V.62 Although the redox potential of Pt(acac)2 is not known for acetonitrile, this value is expected to be similar to that of Pd(acac)2 given that, in water, E° (Pt2+/Pt) is only 0.16 V higher than E° (Pd2+/Pd).63 Thermodynamic data of octyl radicals is not available in the literature, but BPH• is a strong reductant with an oxidation potential of 1.31 V in 50% H2O/ethanol.64 According to this analysis, the much larger quantum yield of Ag formation correlates with a higher thermodynamic tendency of Ag+ to undergo reduction in the nonpolar octane solvent as compared with those of Pd(acac)2 and Pt(acac)2. Equation 2 predicts a maximum φ value of 2 for the formation of ketyl plus alkyl radicals because 3BP* is photogenerated with a quantum yield of 1.48 As in the case of the Cu(oleate)2 photoreduction,39 the fact that φi > 1 for the Ag formation indicates that both BPH• and RH• radicals serve as reductants AgOOR + BRH•/HR• → Ag + ROOH + BP/R

(6)

Thermal conductivity measurements were performed on colloids containing 5 mM Ag, Pd, or Pt; 20 mM BP; and 40 mM OS. Each sample was evaluated at 10, 30, and 50 °C with absolute TC values plotted in Figure 7A, which also includes

■

CONCLUSIONS

■

AUTHOR INFORMATION

Colloidal solutions containing nanometer-sized crystallites of Ag, Pd, and Pt have been generated photochemically in octane using oleoyl sarcosine as a particle stabilizer and BP as a sensitizer. Formation of Ag crystallites occurs with a quantum yield of 1.2, indicating that AgOOR is reduced by both phenyl and ketyl radicals generated during the BP photolysis. Surprisingly, optical signals similar to those observed for Ag clusters in polar solvents are detected during the photoreduction of AgOOR. Solutions containing 0.1 mM Ag and 0.5 mM Pt have thus far remained stable for more than 4 months. On the other hand, Pd particles, prepared by either direct or photosensitized irradiation, precipitate within 1 week. Thermal conductivity measurements on the 5 mM colloids resulted in TC enhancements of up to 10 ± 1%, which is 5 times larger than the values for analogous aqueous systems.47 These observations indicate that efforts to prepare highly concentrated, yet stable, Ag colloids are desirable in order to explore the potential role of the metal particles in PCM systems.

Figure 7. Plots of absolute (A) and relative (B) TC; results from experiments conducted on colloids containing 5 mM of metal, 20 mM BP, and 40 mM OS. The experimentally determined values for octane are in good agreement with reference data.65

Corresponding Author

*E-mail: [email protected].

experimental and literature data for pure octane.65 As shown in the plot, TC of the solvent decreased as T increased, and the experimental results were in agreement with the published values. The opposite trend was noticed for Pd and Pt colloids with TC rising modestly when T increased. Ag colloids exhibited the highest thermal conductivity at the lowest temperatures, but TC decreased slightly with increasing T. In

Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This material is based upon work supported by the Department of Energy under award Number DE-SC0002470. We thank Y. 9248



Article

(33) Wongwises, S.; Daungthongsuk, W. Renewable Sustainable Energy Rev. 2007, 11, 797−817. (34) Maxwell, J. C. Treatise on Electricity and Magnetism; Clarendon Press: Oxford, U.K., 1873. (35) Batchelor, G. K. Annu. Rev. Fluid Mech. 1974, 6, 227−255. (36) Bejan, A.; Kraus, A. D. Heat Transfer Handbook; John Wiley and Sons Inc.: Hoboken, NJ, 2003. (37) Choi, U. S. Enhancing Thermal Conductivity of Fluids with Nanoparticles. In Developments and Applications in Non-Newtonian Flows; Siginer, D. A., Wang, H. P., Eds.; ASME: New York, 1995; Vol. 66, pp 99−105. (38) Khodadai, J. M. Nanoparticle-Enhanced Phase Change Materials (NEPCM) with Great Potential for Improved Thermal Energy Storage. U.S. Patent Application 20090236079, September 24, 2009. (39) Clary, D. R.; Mills, G. J. Phys. Chem. C 2011, 115, 14656− 14663. (40) Fan, L.; Khodadadi, J. M. Renewable Sustainable Energy Rev. 2011, 15, 24−46. (41) Li, D.; Hong, B.; Fang, W.; Guo, Y.; Lin, R. Ind. Eng. Chem. Res. 2010, 49, 1697−1702. (42) This report was prepared as an account of work sponsored by an agency of the United States Government. Neither the United States Government nor any agency thereof, nor any of their employees makes any warranty, express or implied, or assumes any legal liability or responsibility for the accuracy, completeness, or usefulness of any information, apparatus, product, or process disclosed, or represents that its use would not infringe privately owned rights. Reference herein to any specific commercial product, process, or service by trade name, trademark, manufacturer, or otherwise does not necessarily constitute or imply its endorsement, recommendation, or favoring by the Unites States Government or any agency thereof. The views and opinions of authors expressed herein do not necessarily state or reflect those of the United States Government or any agency thereof. (43) Gustafson, S. E. Rev. Sci. Instrum. 1991, 62, 797−804. (44) Heller, H. G.; Langan, J. R. J. Chem. Soc., Perkin Trans 2 1981, 341−343. (45) Creighton, J. A.; Eadon, D. G. J. Chem. Soc., Faraday Trans. 1991, 87, 3881−3891. (46) JCPDS card numbers: Ag, 4-0783; Pd, 46-1043; Pt, 4-0802. (47) Clary, D. R. Ph.D. Dissertation, Auburn University, Auburn, Alabama, November 2011. (48) Gilbert, A.; Baggott, J. Essentials of Molecular Photochemistry; CRC Press: Boca Raton, Fl, 1991; Chapter 7. (49) Cohen, S. G.; Cohen, J. I. J. Am. Chem. Soc. 1967, 89, 164−165. (50) Davidson, R. F.; Lambeth, P. F. J. Chem. Soc., Chem. Commun. 1968, 511−512. (51) Devadoss, C.; Fessenden, R. W. J. Phys. Chem. 1991, 95, 7253− 7260. (52) Winnik, M. A.; Maharaj, U. Macromolecules 1979, 12, 902−905. (53) Lewis, F. D.; Miller, A. M.; Slavi, G. D. Inorg. Chem. 1995, 34, 3173−3181. (54) Korchev, A. S.; Shulyak, T. S.; Slaten, B. L.; Gale, W. F.; Mills, G. J. Phys. Chem. B 2005, 109, 7733−7745. (55) Demeter, A.; Bérces, T. J. Photochem. Photobiol., A 1989, 46, 27−40. (56) (a) Ershov, B. G.; Janata, E.; Henglein, A. J. Phys. Chem. 1993, 97, 339−343. (b) Mostafavi, M.; Dey, G. R.; François, L.; Belloni, J. J. Phys. Chem. A 2002, 106, 10184−10194. (57) Mostafavi, M.; Keghouche, N.; Delcourt, M.-O.; Belloni, J. Chem. Phys. Lett. 1990, 167, 193−197. (58) Gaddy, G. A.; McLain, J. L.; Korchev, A. S.; Slaten, B. L.; Mills, G. J. Phys. Chem. B 2004, 108, 14858−14865. (59) Linnert, T.; Mulvaney, P.; Henglein, A. J. Phys. Chem. 1993, 97, 679−682. (60) (a) Mulvaney, P.; Linnert, T.; Henglein, A. J. Phys. Chem. 1991, 95, 7843−7846. (b) Henglein, A.; Mulvaney, P.; Holzwarth, A.; Sosebee, T. E. Ber. Bunsen-Ges. Phys. Chem. 1992, 96, 754−759.

Berta of the Georgia Institute of Technology for the timely help with electron diffraction experiments.

■

REFERENCES

(1) Sakamoto, M.; Majima, T. J. Photochem. Photobiol., C 2009, 10, 33−56. (2) Yonezawa, Y.; Sato, T.; Ohno, M.; Hada, H. J. Chem. Soc., Faraday Trans. 1 1987, 83, 1559−1567. (3) (a) Harada, M.; Saijo, K.; Sakamoto, N.; Ito, K. J. Colloid Interface Sci. 2010, 343, 423−432. (b) Einaga, H.; Harada, M. Langmuir 2005, 21, 2578−2584. (4) (a) Kometani, N.; Doi, H.; Asami, K.; Yonezawa, Y. Phys. Chem. Chem. Phys. 2002, 4, 5142−5147. (b) Harada, M.; Saijo, K.; Sakamoto, N.; Ito, K. J. Colloid Interface Sci. 2008, 325, 1−6. (c) Henglein, A. Chem. Mater. 1998, 10, 444−450. (5) (a) Kometani, N.; Doi, H.; Asami, K.; Yonezawa, Y. Phys. Chem. Chem. Phys. 2002, 4, 5142−5147. (b) Harada, M.; Saijo, K.; Sakamoto, N.; Ito, K. J. Colloid Interface Sci. 2008, 325, 1−6. (c) Henglein, A. Chem. Mater. 1998, 10, 444−450. (6) Li, H.; Duan, X.; Liu, G.; Liu, X. J. Mater. Sci. 2008, 43, 1669− 1676. (7) Cozzoli, P. D.; Comparelli, R.; Fanizza, E.; Curri, M. L.; Agostiano, A.; Laub, D. J. Am. Chem. Soc. 2004, 126, 3868−3879. (8) Giuffrida, S.; Ventimiglia, G.; Petralia, S.; Conoci, S.; Sortino, S. Inorg. Chem. 2006, 45, 508−510. (9) Harada, M.; Einaga, H. Langmuir 2006, 22, 2371−2377. (10) Harada, M.; Inada, Y. Langmuir 2009, 25, 6049−6061. (11) Murakata, T.; Higashi, Y.; Yasui, N.; Higuchi, T.; Sato, S. J. Chem. Eng. Jpn. 2002, 35, 1270−1276. (12) Nath, S.; Jana, S.; Pradhan, M.; Pal, T. J. Colloid Interface Sci. 2010, 341, 333−352. (13) Brust, M.; Walker, J. M.; Bethell, D.; Schiffrin, D. J.; Whyman, R. J. Chem. Soc., Chem. Commun. 1994, 801−802. (14) Esumi, K.; Shiratori, M.; Ishizuka, H.; Tano, T.; Torigoe, K.; Meguro, K. Langmuir 1991, 7, 457−459. (15) Manna, A.; Imae, T.; Iida, M.; Hisamatsu, N. Langmuir 2001, 17, 6000−6004. (16) Nath, S.; Praharaj, S.; Panigrahi, S.; Ghosh, S. K.; Kundu, S.; Basu, S.; Pal, T. Langmuir 2005, 21, 10405−10408. (17) Song, X.; Sun, S.; Zhang, W.; Yin, Z. J. Colloid Interface Sci. 2004, 273, 463−469. (18) Xun, F.; Wei, Y.; Yusheng, L.; Debao, W.; Huaqiang, S.; Fengyuan, Y. J. Dispersion Sci. Technol. 2005, 26, 575−580. (19) Lala, N.; Lalbegi, S. P.; Adyanthaya, S. D.; Sastry, M. Langmuir 2001, 17, 3766−3768. (20) Zeiri, L.; Efrima, S. J. Phys. Chem. 1992, 96, 5908−5917. (21) Vorobyova, S. A.; Lesnikovich, A. I.; Sobal, N. S. Colloids Surf., A 1999, 152, 375−379. (22) Kang, S. Y.; Kim, K. Langmuir 1998, 14, 226−230. (23) Yang, J.; Lee, J. Y.; Chen, L. X.; Too, H. P. J. Phys. Chem. B 2005, 105, 5468−5472. (24) Bö nnemann, H.; Bladergroen, B.; Linkov, V. M. Appl. Organomet. Chem. 2005, 19, 768−773. (25) Yamamoto, M.; Kashiwagi, Y.; Nakamoto, M. Langmuir 2006, 22, 8581−8586. (26) Rao, C. R. K.; Trivedi, D. C. Mater. Chem. Phys. 2006, 99, 354− 360. (27) Hiramatsu, H.; Osterloh, F. E. Chem. Mater. 2004, 16, 2509− 2511. (28) Esumi, K.; Itakura, T.; Torigoe, K. Colloids Surf., A 1994, 82, 111−113. (29) Wadkar, M. M.; Chaudhari, V. R.; Haram, S. K. J. Phys. Chem. B 2006, 110, 20889−20894. (30) Taleb, A.; Petit, C.; Pileni, M. P. Chem. Mater. 1997, 9, 950− 959. (31) Li, Y.; Wu, Y.; Ong, B. S. J. Am. Chem. Soc. 2005, 127, 3266− 3267. (32) Das, S. K.; Choi, U. S.; Patel, H. E. Heat Transfer Eng. 2006, 27, 3−19. 9249



Article

(61) (a) Song, L.; Trogler, C. Angew. Chem., Int. Ed. Engl. 1992, 31, 770−772. (b) Pavlishchuk, V. V.; Addison, A. W. Inorg. Chim. Acta 2000, 298, 97−102. (62) Murray, R. W.; Hiller, K. J. Anal. Chem. 1967, 39, 1221−1229. (63) Lide, D. R., Ed. CRC Handbook of Chemistry and Physics, 73rd ed.; CRC Press, Inc.: Boca Raton, FL, 1992. (64) Canonica, S.; Hellrung, B.; Wirz, J. J. Phys. Chem. A 2000, 104, 1226−1232. (65) Vargaftik, N. B., Filippov, L. P., Tarzimanov, A. A., Totskii, E. E., Eds. Handbook of Thermal Conductivity of Liquids and Gases; CRC Press: Boca Raton, FL, 1994. (66) Cverna, F. Thermal Properties of Metals; ASM International: Materials Park, OH, 2002; Chapter 4. (67) Eapen, J.; Rusconi, R.; Piazza, R.; Yip, S. J. Heat Transfer 2010, 132, 102402.

9250


Photochemical Generation of Ag, Pd, and Pt Particles in Octane - The

Recommend Documents