Separation and Puri?cation Technology43(2005)
9–15
Gold recovery from parts-per-trillion-level aqueous solutions
by a nanostructured Mn2O3adsorbent
H.Koyanaka a,1,K.Takeuchi b,C.-K.Loong c,?
a CRMD/CNRS,Orleans University,45071Orleans,Cedex2,France
b Tokyo University of Science,Oshamanbe,Hokkaido049-3514,Japan
c Argonne National Laboratory,Argonne,IL60439,USA
Received7June2004;received in revised form15September2004;accepted15September2004
Abstract
We introduce a nanostructured Mn2O3-based adsorbent and a low-cost method that is capable of extracting gold,in the form of metallic nano-to-micrometer-size particles,from down to sub-ppm-level aqueous solutions including seawaters with high yield,good selectivity and recyclability,and environmental benignity.The Mn2O3adsorbent,prepared by low-temperature synthesis followed by acid treatments,exhibits a mass-fractal-like morphology of agglomerated nanometer-size crystallite grains covered with active protonated sites and a surface area of about120m2/g.For aqueous solutions containing～100ppm gold,a yield of70mg of gold/g of adsorbent was achieved.A Langmuir-type adsorption isotherm was observed,showing a rapid uptake of gold.The method is well suited to gold recovery from very dilute solutions. For example,over95%of the gold was recovered from seawater samples containing0.1and1ppm of added gold.A pilot study of seawater with1ppt added gold showed similar results.The protonation–suspension–?ltration–washing process can be recycled without noticeable degradation in yield.
?2004Published by Elsevier B.V.
Keywords:Gold recovery;Manganese oxide adsorbent;Gold in seawater;[Au(I)Cl2]?and[Au(III)Cl4]?complexes
1.Introduction
Abundance of gold in the Earth’s crust varies between～10 parts per million(ppm)[1]in commercial ore and～10parts per trillion(ppt)in seawater[2–4].Any method attempting to recover gold from natural waters has to be extremely ef-?cient in order to be economically viable.Dissolved gold ions in chlorinated aqueous solutions form[Au(I)Cl2]?and [Au(III)Cl4]?complexes.[AuCl2]?is more stable in di-lute solutions of nearly neutral pH such as seawater,al-though questions concerning its stability relative to that of the AuOH(H2O)0species under oxic conditions have been raised[2,5].In any case,slight increase of acidity or chlori-nation will stabilize all the species to[AuCl2]?complexes ?Corresponding author.Tel.:+16302525596;fax:+16302524163.
E-mail address:ckloong@https://www.doczj.com/doc/2017688349.html,(C.-K.Loong).
1Permanent address:Dannoharu900-103,Oita-shi,870-1124,Japan.[6].Employment of adsorbents to recover gold from natural waters,due to the extreme dilution,requires effective sus-pension of the adsorbing particles with maximally allowed, chemically active sites that target speci?cally[AuCl2]?com-plexes and exceedingly high rate of Au(I)to Au(0)reduc-tion.Previous methods of gold recovery were either feasi-ble only to specially processed solutions containing rather high gold concentrations or deemed impractical due to low yields,operational dif?culties or environmental concerns. Such adsorbents include silica-polymer composites[7],sur-factants[8],pyrite,goethite,and birnessite[9,10]applicable to acidic hydrochloric solutions(pH<4)or seawater contain-ing1–130ppm of gold,activated carbons or ion-exchange resins for recovering gold from cyanide-leached solutions in the mining industry[11,12],and proposed utilization of algae [13]and micro-organisms[14].Hence an adsorbent capable of recovering gold from natural waters down to ppt levels with high ef?cacy and minimal environmental impact is highly de-
1383-5866/$–see front matter?2004Published by Elsevier B.V. doi:10.1016/j.seppur.2004.09.005
10H.Koyanaka et al./Separation and Puri?cation Technology43(2005)9–15
sirable.Furthermore,cost consideration demands a relatively
simple method of synthesis and chemical processing of the
adsorbent.
Our contemplation of a suitable manganese-oxide ad-
sorbent was motivated by several important properties
common to manganese oxide/hydroxide minerals:(i)the
ubiquitous presence in the ocean?oor,soils,and sed-
iments,which imply environmental benignity[15];(ii)
the high cation-exchange capacity;and(iii)the available
Mn(II),Mn(III)and Mn(IV)states that are amenable to
oxidation–reduction reactions.However,the Mn(IV)oxida-
tion state in MnO2-type compounds is known to promote
dissolution of gold in hydrochloric solutions.For example,
birnessite(Na,K)Mn7O14·2.8H2O)is not effective in ad-sorbing gold[9].We hypothesize that the Mn(III)state in
an Mn2O3-type adsorbent with a nanostructured architec-
ture capable of controlled association/dissociation of pro-
tonic species on/off the particle surfaces is essential to ef-
?cient gold extraction.Such adsorbent was realized by a
low-temperature synthesis of a nanostructured Mn2O3pow-
der,followed by protonation of the particle surfaces by acid
treatments.The Mn2O3adsorbents were systematically ap-
plied to specimens of distilled water and seawater contain-
ing1–100ppm and0.1–1ppm of added gold,respectively,
through a suspension–?ltration–washing procedure.Encour-
aged by the results,we studied a seawater sample containing
1ppt added gold.
2.Experimental
2.1.Low-temperature synthesis and acid treatment
An MnCO3powder(Wako,99%purity)as the start-
ing material was heated using an electric furnace at230?C
in air for4.5h.An acid treatment began with adding the
powder into in a0.5mol/L of nitric or hydrochloric acid
whiling stirring(using a magnetic stirrer)for a chosen pe-
riod of suspension time.Afterward the suspended particles
were?ltered from the solution using a glass?lter(0.2?m
mesh),followed by washing with distilled water.This?rst-
stage processing with a15min suspension time for an initial
acid treatment enabled the growth of nanometer-size crys-
talline Mn2O3grains on the surfaces of micron-size crys-
talline MnCO3particles.Subsequent acid treatments,typ-
ically for1h suspension each,allowed the removal of the
MnCO3component and the self-assembly of adsorbed H+
ions(protons)on the Mn2O3nanoparticles.This second
acid treatment resulted in a substantial increase of the pop-
ulation of the protonated sites.Furthermore,the remain-
ing MnCl2or MnNO3in the solution after an acid treat-
ment and?ltration of the adsorbent can readily be con-
verted to useful high surface area MnO2powder by adding
an appropriate amount of KMnO4to the solution.This
reduces the amount of waste and negative environmental
impacts.2.2.Characterization and chemical analysis
The microstructure and crystal phases were character-ized by small-angle neutron scattering(SANS),neutron powder diffraction(NPD)and synchrotron-radiation X-ray diffraction(SRXD),and the morphology by scanning(SEM) and transmission(TEM)electron microscopy.Inductively coupled plasma atomic-emission/mass-spectroscopy(ICP-AES/MS),energy dispersive X-ray(EDX),and X-ray pho-toelectron(XPS)spectroscopies were employed for chemi-cal analyses.Neutron studies were performed at the Intense Pulsed Neutron Source of Argonne National Laboratory.The SRXD experiment was carried out at the synchrotron X-ray source,SPring8(Hyogo,Japan).ICP-MS measurements were conducted independently by the Centre National de la Recherche Scienti?que Service Central d’Analyse in Or-leans,France.SEM,TEM,ICP-AES,EDX,XPS measure-ments were performed using standard equipments.
2.3.Seawater experiments
Care was used to collect unpolluted near-surface seawa-ter from Beppu Bay,Japan and to store it in acid-cleaned polyethylene bottles.For the0.1and1ppm added gold con-centrations,the goal was to determine the amount of gold ad-sorption over different suspension times but not the residual concentration of gold in the solution after achieving equilib-rium in saturated adsorption,hence using0.5g of adsorbent in1L of seawater was suf?cient.In the case of1ppt added gold in seawater,5g of adsorbent was suspended in10L of seawater.Here,we did not intend to assess the natural gold content in the seawater but to measure the elution curve of the recovered gold by the adsorbent.The expected ppb level of gold,if present in the solution obtained by washing the?ltered adsorbent with～10mL of0.5mol/L hydrochloric acid,was well within the capability of ICP-MS detection.The ultra-pure water(Wako)and ultra-pure hydrochloric acid(Wako) used in the experiment were found to contain gold,if any,at
a level below the detection limit.
3.Results and discussion
3.1.Structures at micrometer to nanometer scale
We?rst determined the crystal structures of the compos-ite of micrometer-size MnCO3and nanometer-size Mn2O3, obtained from the?rst-stage synthesis.Fig.1shows the re-?ned NPD and SRXD data for the crystal structures which contain4.6wt.%of cubic-Mn2O3[16]and95.4wt.%of rhombohedral-MnCO3[17].No other phases were observed other than an amorphous-like component as evidenced in the residual intensity,presumably originated from atomic dis-order at the interfaces of the particles.The microstructure of the composite was characterized by SANS.The observed intensity pro?le arises from the differences in the neutron
H.Koyanaka et al./Separation and Puri?cation Technology 43(2005)9–15
11
Fig.1.Rietveld re?nements of the room-temperature NPD pattern (dots:data,line:calculated pro?le)of the as-sintered material treated with acid for 15min.The crystal phases include MnCO 3and Mn 2O 3,of which the Bragg-re?ection positions are denoted by the lower and upper tick marks,respectively.The indices of some re?ections are given.The Bragg-peak intensity of the Mn 2O 3phase,albeit tiny,can clearly be seen by neutrons (inset)and synchrotron X-rays (thick line above the tick marks).The absence of the (006)re?ection of the MnCO 3phase in the X-ray data is due to its vanishingly small structure factor.
scattering-length densities of MnCO 3and Mn 2O 3particles and the voids over the (reciprocal)length scale τin terms of the wavevectors Q ,where Q =2π/τ.The SANS pro?le in Fig.2shows a major component of large MnCO 3particles
(at Q <0.006?A
?1)and a minor component of smaller Mn 2O 3particles (a broad shoulder at Q ≈0.02?A
?1).A slope of the curve of slightly higher than ?4(the Porod value)at high Q indicates the rough interfaces of the particles [18].
The neutron and X-ray diffraction studies of the average structure of the samples were complemented by microscopy measurements in the real space.As shown in Fig.3a,TEM con?rmed the morphology of plate-like Mn 2O 3nanoparticles on the surfaces of large MnCO 3particles.The surface area of the material before and after the second acid treatment,estimated from a nitrogen adsorption isotherm measurement and a (Brunauer–Emmett–Teller)BET analysis,was found to be ～90and ～120m 2/g,respectively.Subsequent acid treat-ments did not increase the surface area.Both TEM and BET measurements indicated a microporosity of the adsorbent with an average pore diameter of about 10
nm.
Fig.2.SANS pro?les of the samples obtained after sintering and acid treat-ments.The errors are comparable to the size of the symbols except for
Q >0.4?A
?1where the errors are larger.The slopes of the curves at high Q are slightly higher than ?4.The thick curve represents a ?t of the data (3times acid treatment)with a mass-fractal model.The SANS pro?le of a powder undergone two additional acid-treatments (Fig.2)shows a large decrease of the inten-sity at low-Q and the appearance of a broad maximum at
Q >～0.01?A
?1,corresponding to the removal of the MnCO 3and aggregation of the Mn 2O 3nanoparticles.The nature of the aggregation can be described satisfactorily by a mass-fractal geometry,namely,a fractal dimension of 3.02±0.01and a mean particle size of 10.3±0.03nm with a log-normal distribution of sizes characterized by a standard deviation of 1.65±0.01[19,20].The SANS result is corroborated by the TEM data (see Fig.3b).Additional NPD and neutron spectroscopic studies con?rm that protonation of the Mn 2O 3nanoparticles via acid treatments introduced distortions of the crystal lattice,atomic disorder,and defects.Since the proton coverage is large,the modi?cation of the crystal structure and stoichiometry is substantial.For convenient,we refer the acid-treated powders as Mn 2O 3adsorbent but do not imply strictly the crystal structure and stoichiometry of Mn 2O 3[16].3.2.Gold recovery from 1to 100ppm aqueous solutions We ?nd that two additional acid-treatments (total suspen-sion time 2h)of the initially prepared Mn 2O 3powder pro-duced very effective gold adsorbents.The gold adsorption experiments were conducted by suspending 1g of
adsorbent
Fig.3.TEM imagine of the sample:(a)the large MnCO 3grains and small,nm-size plate-like Mn 2O 3are observed;(b)after 3times of acid treatment only an aggregate of the Mn 2O 3crystallites in microporous geometry is evident.The featureless background near the edges is from the carbon grid of the sample holder.
12H.Koyanaka et al./Separation and Puri?cation Technology 43(2005)9–15
in 3L of aqueous solutions containing 1,20,30,50,and 100ppm of gold that were prepared by mixing a standard HAuCl 4solution with distilled water while maintaining a pH of ～6by adding appropriate amounts of a 1mol/L aqueous NaOH solution.A 10mL batch of the ?ltered solution was collected at regular time intervals,from which the gold con-centrations were determined by ICP-AES.Fig.4a shows a rapid decrease of the gold concentration in the 1-ppm solu-tion as a function of suspension time.The obtained
adsorption
Fig.4.(a)Decrease of the gold concentration in the solutions as a function of suspension time of the adsorbent.(b)Yield in units of mg of gold/g adsor-bent vs.the gold concentration of the solutions at equilibrium of saturated adsorption.The initial gold concentrations in units of ppm are given next to the data points.(c)Elution curve of gold recovery from the adsorbent (5g)after suspension in 10L of seawater containing 1ppt added gold by washing with 0.5mol/L ultra-pure hydrochloric acid.The area in a grey box repre-sents the amount of gold in each elution from ICP-MS measurements.The hatched boxes are the error bars of the measurements.The total recovered gold of about 19ng (with an uncertainty of ～35%)is in reasonable agree-ment with the 10ng added gold in 10L plus the natural content of gold in the seawater
sample.
Fig.5.Decline of Au concentration as a function of suspension time for a virgin and once and twice regenerated adsorbent;0.3g of adsorbent was added in a 3L aqueous solution containing 5ppm of gold while maintaining a pH of 5.0–6.0.
isotherm (Fig.4b),in terms of yield versus gold concentra-tion in the solutions at equilibrium of saturated adsorption,shows:(i)an achieved yield of 70mg of gold/g of adsorbent that supersedes those of other methods of gold extraction from aqueous solutions [7–10];and (ii)a sharp rise from the origin typical of the Langmuir-type isotherm [21].The rapid takeoff of isotherm demonstrates the high ef?ciency of this adsorbent,hence the applicability to very dilute solutions.In approaching to the maximal yield of ～70mg gold/g adsor-bent,apparently,on the average each nanoparticle is covered by a monolayer of gold.Re-activation of the adsorbent is sim-ple.Several washings of the ?ltered adsorbent with 10mL of 0.5mol/L hydrochloric acid dissolve all the adsorbed gold and re-protonate the active sites at the same time.After rinsing with distilled water,the adsorbent is ready for re-using with nearly the original ef?cacy.SANS and TEM con?rmed the preservation of the nanostructured morphology in the gold-adsorbed and re-activated adsorbents.Fig.5illustrates the typical performance of a reused adsorbent.About 0.3g of adsorbent was suspended in 3L of 5ppm gold aqueous so-lution while maintaining a pH of 5.0–6.0.Fig.5shows the decline of Au concentration for a virgin and once and twice regenerated adsorbent as a function of suspension time.The adsorption rates are very similar.
3.3.Gold recovery from 0.1and 1ppm and 1ppt seawater
For the seawater sample with 0.1and 1ppm added gold,we ?nd that the adsorbent was equally effective (over 95%re-covery rate)with the exception that up to 18h of suspension was required for full recovery of the gold.Apparently,the presence of high-concentration cations such as Na +,Mg 2+,and Ca 2+within the bilayer surrounding a suspended particle interferes the proton release into the solution and retards the reduction of the AuCl 2?complexes (see Section 3.4below).The estimated contents (in ppm)of these ions in the adsorbent by ICP-AES were Ca (～200)>Na (48)>Mg (7.8).Compar-ing with the nominal content of Na (10770)>Ca (412)>Mg
H.Koyanaka et al./Separation and Puri?cation Technology 43(2005)9–1513
(190)in seawater [22],the selectivity of gold against other ions is good.In the case of 1ppt added gold in seawater,the main purpose was to measure the elution curve of the recovered gold by the adsorbent.The elution was obtained by washing the ?ltered adsorbent with ～10mL of 0.5mol/L hydrochloric acid.Fig.4c show the elution curve of gold recovered from the 1ppt gold-added seawater.The total re-covered gold of about 19ng (with an uncertainty of ～35%)is in reasonable agreement with the 10ng added gold in 10L plus the natural content of gold in the seawater sample.3.4.Adsorbed metallic gold particles
The adsorbed gold on the Mn 2O 3particles were charac-terized by TEM,SEM,XPS and EDX measurements.XPS measurements identi?ed a binding energy of Au 4f 7/2orbital of 84eV which is consistent with the metallic Au(0)state of the adsorbed gold.If the 3+oxidation state is present as it would be the case for a [Au(III)Cl 4]?complex,a binding energy of 85.7–87.1eV is expected [23,24].Figs.6a and b show that initially,polydisperse nanometer-size gold parti-cles were formed.These particles subsequently grew to sub-micron sizes,which was con?rmed by SEM and concurrent EDX characterization (Fig.7a and
7b).
Fig.6.High-resolution (a)and low-resolution (b)TEM images of gold par-ticles (dark)adsorbed on the Mn 2O 3adsorbent (light).
The data demonstrate the effectiveness of our Mn 2O 3ad-sorbent in recovering gold from distilled water and seawater down to ～1ppt concentrations.The mechanism begins with the deposition of AuCl 2?complexes on the surfaces of the positively charged Mn 2O 3nanoparticles.Protons are subse-quently released,as evidenced by an initial decrease in the pH of the solution when the adsorbent was added,and electron transfers take place for decomposition of the AuCl 2?com-plexes into metallic gold via the well-known reaction [6,25]:3AuCl 2?→2Au(s)+AuCl 4?+2Cl ?.
(1)
Nanometer-size seeds of metallic gold are formed initially,many of them grow to micron-size particles subsequently by a self-catalyzed reaction as previously studied by Gammons and coworkers [6].However,the present result is rather surprising.First,the proton-activated,mass-fractal-like nanostructure of the powder drastically enhances the reactivity,hence an ultrahigh ef?ciency of gold recovery,to an unprecedented level.Traditional electrochemical approaches,even with the application of electrodes,can-not produce the high adsorption rate in such very dilute solutions.Any theory of gold reduction via redox reactions (e.g.,Mn(III)→Mn(IV)and Au(III and I)→Au(0)
with
Fig.7.SEM image (a)showing the grown gold particles (with lighter color)on the adsorbent,and the correspondent EDX image (b)showing the distri-bution of the large and small gold particles.
14H.Koyanaka et al./Separation and Puri?cation Technology43(2005)9–15
redox potentials of0.54V at pH=8and1.002–1.154V, respectively[26]),if present additionally,has to take into account of the interfacial interactions between the AuCl2?complexes and the protonated Mn2O3layers at a nanoscale.Perhaps,the ultrahigh ef?ciency is made possible by the replenishment of Au(III)Cl2?species in Eq.(1)via oxidation of Mn(III)of Mn2O3to form MnO2. However,the partial conversion of Mn(III)to Mn(IV) cannot be con?rmed by XPS measurement because of the overlapping binding energies of these two states from the 2p3/2orbital.Secondly,the pro?ciency of dissolving all the metallic gold by washing the Mn2O3adsorbent with only dilute(～0.5mol/L)hydrochloric solutions,as consistently shown by the ICP measurements of the gold contents in the adsorbent and in the extracted solutions,is unusual. The extracted solutions display a gold color of dissolved Au complexes as in the standard HAuCl4solution as opposed to the red color of metallic gold colloids.Normally, special acids such as aqua regina are required to dissolve metallic gold.We speculate a mechanism of dissolving the adsorbed gold via a crossover of the redox potentials of Mn(IV)→Mn(II)and Au(0)→Au(III)in the reaction: 3MnO2+12H++2Au+8Cl?→3Mn2++6H2O+2AuCl4?, which presumably can only occur at a nanometer scale.
4.Conclusions
We summarize the synthesis,processing,and characteri-zation of the Mn2O3-based adsorbent and the gold recovery as follows:
(1)A composite consisting of crystallite,micron-size
MnCO3and nm-size Mn2O3particles was synthesized by a low-temperature method.
(2)Acid treatments of the composite removed the MnCO3
component and protonated the Mn2O3nanoparticle sur-faces.This process produced an adsorbent having a fractal-like morphology with a surface area of120m2/g.
(3)The adsorption isotherm,elution curves and the prod-
uct of gold recovery from aqueous solutions and seawater samples containing gold through a suspen-sion/?ltration/washing procedure were described.Chem-ical analysis showed that for20–100ppm gold concen-trations where the equilibrium saturated concentration could be measured,a maximal yield of70mg gold/g ad-sorbent was achieved.For lower concentrations down to～1ppt level,elution curves showed that essentially all the gold could be recovered.The adsorbed gold was metallic Au(0)particles,which could subsequently be extracted by dissolution in dilute HCl solutions via wash-ing.
(4)The adsorbent can be reused with sustainable ef?ciency
simply by washings with dilute HCl solutions.
(5)A conjecture of the unique capabilities of this method,
namely,ultrahigh ef?ciency of gold recovery from very
dilute solutions and the solubility of the extracted metal-lic gold particles by dilute HCl solutions was given,em-phasizing the redox character of the nanoscaled man-ganese oxide adsorbent.
The high yield and selectivity,excellent ef?ciency,and recyclable usage of this gold adsorbent in an environmen-tally benign manner clearly demonstrated the feasibility of recovering gold from aqueous solutions and seawater down to ppb levels.The tantalizing result of the1ppt added gold in seawater might merit an industrially scale-up evaluation for economical viability of the method.To reduce the cost of initial investment,testing can be integrated with the ex-isted infrastructure of the desalination plants and power re-actors that use seawater as coolant[27],as previously demon-strated for the case of lithium collection from seawater using a manganese-spinel adsorbent[28].Moreover,preliminary tests showed that our Mn2O3adsorbent(0.5g)could recover 60%of silver and70%of palladium from40ppm,1L aque-ous solutions.Post-sintering can control the nanometer size distribution of the Ag and Pd particles.Therefore,in addition to preparing ef?cient Mn2O3adsorbents for extract useful metals from aqueous solutions,our synthesis method can in principle produce useful Mn2O3-based catalysts supporting different metallic nanoparticles.
Acknowledgement
We thank N.Koura,J.W.Richardson,Jr.,V.Komenko and P.Azais for the helpful discussions.The assistance in sample characterization from C.Clinard and T.Cacciaguerra with TEM,E.Veron and A.Uchida with SEM,P.Baillif with ICP-AES,D.Wozniak and L.Guo with SANS,and S.Kohara with SRXD is gratefully acknowledged.HK thanks Des Bourses 2003du Gouvernement Francais for the?nancial support and M.-L.Saboungi for the kind hospitality during the stay at CRMD.Work performed at Argonne National Laboratory is supported by the U.S.DOE-Basic Energy Science under the contract No.W-31-109-ENG-38.
References
[1]For aqueous solutions,the units of solute concentration,ppm,ppb
and ppt are equivalent to mg/kg,?g/kg and ng/kg,respectively.
[2]K.K.Falkner,J.M.Edmond,Earth Planet Sci.Lett.98(1990)208.
[3]J.B.McHugh,J.Geochem.Expl.30(1988)85.
[4]M.Koide,V.Hodge,E.D.Goldberg,Appl.Geochem.3(1988)237.
[5]D.Vlassopoulos,S.A.Wood,Geochim.Cosmochim.Acta54(1990)
3.
[6]C.H.Gammons,Y.Yu,A.E.Williams-Jones,Geochim.Cosmochim.
Acta61(1997)1971.
[7]K.G.Neoh,K.K.Tan,P.L.Goh,S.W.Huang,E.T.Kang,K.L.Tan,
Polymer40(1999)887.
[8]T.Kinoshita,S.Akita,S.Ozawa,S.Nii,F.Kawaizumi,K.Takahashi,
J.Min.Mater.Char.Eng.2(2003)71.
[9]Y.Ran,J.Fu,A.W.Rate,R.J.Gilkes,Chem.Geo.185(2002)33.
H.Koyanaka et al./Separation and Puri?cation Technology43(2005)9–1515
[10]M.A.A.Schoonen,N.S.Fisher,M.Wente,Geochim.Cosmochim.
Acta56(1992)1801.
[11]N.Tsuchida,D.M.Muir,Metall.Trans.17B(1986)523.
[12]C.P.Gomes,M.F.Almeida,J.M.Loureiro,Sep.Purif.Technol.24
(2001)35.
[13]T.Kuwabara,H.Hirama,A.Yazawa,J.Mining Mater.Process.Inst.
Jpn.110(1994)15.
[14]F.Relth,AusIMM Bull.Nov/Dec(2003)60.
[15]J.E.Post,Proc.Natl.Acad.Sci.U.S.A.96(1999)3447.
[16]S.Geller,J.Chem.Phys.33(1960)676.
[17]D.L.Grap,Am.Miner.46(1961)1283.
[18]G.Porod,in:O.Glatter,O.Kratky(Eds.),Small Angle X-ray Scat-
tering,Academic Press,New York,1982,p.17.
[19]S.K.Sinha,T.Freltoft,J.Kjems,in:F.Family,https://www.doczj.com/doc/2017688349.html,ndau(Eds.),
Kinetics of Aggregation and Gelation,Elsevier Science,Amsterdam, 1984,p.87.
[20]C.-K.Loong,P.Thiyagarajan,J.W.Richardson,M.Ozawa,S.
Suzuki,J.Catal.171(1997)498.
[21]The adsorption isotherm of Fig.4b,expressed in yield versus residual
gold concentration at saturated adsorption,does not include data
points of initial concentrations below20ppm because,as the rapid adsorption of all the gold for the1ppm gold solution shown in Fig.4a,the volume of the solution required for such measurement is beyond the scale achievable in a laboratory.
[22]H.U.Sverdrup,M.W.Johnson,R.H.Fleming,The Oceans:Their
Physics,Chemistry and General Biology,Prentice-Hall,New York, 1942.
[23]D.Briggs,M.P.Seah,Practical Surface Analysis,Wiley,New York,
1996.
[24]C.D.Wagner,W.M.Riggs,L.E.Davis,J.F.Moulder,in:G.E.Muilen-
berg(Ed.),Handbook of X-ray Photoelectron Spectroscopy,Perking-Elmer,Eden Prairie,Minnesota,1979.
[25]J.J.Lingane,J.Electroanal.Chem.4(1962)332.
[26]M.Pourbaix,Atlas of Electrochemical Equilibria in Aqueous Solu-
tions,National Association of Corrosion Engineers,Houston,TX, 1974.
[27]The intake of seawater as coolant for a1GW power plant or as
feedwater to a desalination plant is typically2000metric tons/min.
[28]H.Koyanaka,Y.Koyanaka,Y.Numata,T.Wakamatsu,J.Mining
Mater.Process.Inst.Jpn.113(1997)275.
实时频谱分析仪（RTSA），这是基于快速傅利叶（FFT）的仪表，可以实时捕获各种瞬态信号，同时在时域、频域及调制域对信号进行全面分析，满足现代测试的需求。 一、实时频谱分析仪的工作原理 在存在被测信号的有限时间内提取信号的全部频谱信息进行分析并显示其结果的仪器主要用于分析持续时间很短的非重复性平稳随机过程和暂态过程，也能分析40兆赫以下的低频和极低频连续信号，能显示幅度和相位。 傅里叶分析仪是实时式频谱分析仪，其基本工作原理是把被分析的模拟信号经模数变换电路变换成数字信号后，加到数字滤波器进行傅里叶分析；由中央处理器控制的正交型数字本地振荡器产生按正弦律变化和按余弦律变化的数字本振信号，也加到数字滤波器与被测信号作傅里叶分析。正交型数字式本振是扫频振荡器，当其频率与被测信号中的频率相同时就有输出，经积分处理后得出分析结果供示波管显示频谱图形。正交型本振用正弦和余弦信号得到的分析结果是复数，可以换算成幅度和相位。分析结果也可送到打印绘图仪或通过标准接口与计算机相连。 二、实时频谱分析仪中的数字信号处理技术 1. IF 数字转换器 一般会数字化以中间频率(IF)为中心的一个频段。这个频段或跨度是可以进行实时分析的最宽的频率范围。在高IF 上进行数字转换、而不是在DC 或基带上进行数字转换，具有多种信号处理优势(杂散性能、DC抑制、动态范围等)，但如果直接处理，可能要求额外的计算进行滤波和分析。 2. 采样 内奎斯特定理指出，对基带信号，只需以等于感兴趣的最高频率两倍的速率取样 3. 具有数字采集的系统中触发 能够以数字方式表示和处理信号，并配以大的内存容量，可以捕获触发前及触发后发生的事件。数字采集系统采用模数转换器(ADC)，在深内存中填充接收的信号时戳。从概念上说，新样点连续输送到内存中，最老的样点将离开内存。
幻想战姬竞技场剑系4大战姬推荐 幻想战姬竞技场剑系4大战姬推荐，推荐幻想战姬竞技场剑系4大战姬。希望这篇幻想战姬竞技场剑系4大战姬推荐，能帮助到各位正在玩幻想战姬的玩家朋友们！ 剑系第一名：花嫁妲己 排第一的毫无疑问是花嫁妲己，从属性上来看，姬65的花嫁接近4W血，1600的攻击。从技能上来说，后排主动高倍率全体AOE，被动贯通回能量。从卡面来看，呆萌的新娘子，你还要啥自行车? 这么完美的卡你有什么不练她的理由? 我想很多人都有打死对面三个后被花嫁一个AOE清场的经历，所以，这卡的评价，中低端局有个作为核心卡的能力，高端局则绝对是目前最强辅助。 剑系第二名：孙悟空 首先，大圣的血量足以支撑其撑在前排。并且前排主动技能贯通伤害的倍率也不错，大圣在前排，对对面除妲己以外的盾系职业来说都是毁灭性的打击。唯一的不足就是，大圣的后排技能略逗，放后排没什么意义。
剑系第三名：哪吒 首先我说要，哪吒后排的技能相当不错。在PVE中对很多难啃的BOSS有奇效，许多人都是靠天女和大腿的哪吒过的老牛，而且哪吒的攻击力成长也非常不错。但哪吒这卡有几个不协调的 缺点，首先，作为前排，血过少，不是很能抗，而且作为前排，贯通技能的倍率略感人，作为后排，虽然技能的眩晕3回合对PVP来说十分不错，发动实在太慢，很可能在发动前就会被打飞。
剑系第四名：钟馗姬 说实话，馗姬更适合推图，血量一般，前排AOE技能的倍率也一般。攻击虽然不错，但站不住的话就意义不大了，JJC中用的人也不算多。 最后说几张福袋剑卡
狂欢卡莉： 就是弱化版的后排钟馗姬，PVP价值不大。 心愿土特产： 技能相当优秀，奈何本身属性实在不行，本来是唯一可以痛打盾妲己的存在，奈何从属性上来看却不能对盾妲己造成什么实质的威胁。 华阳： 只能说技能看上很美，但是有两点致命缺点一、横扫回能量只回后排二、和贯通回能量不叠加。所以，实际来说，华阳很鸡肋。 另外，推荐非洲战神之一的赵萌萌，首充即送，初始相阶，成长却意外的还不错，前排技能倍率也很不错，是各位非洲大草原人民最忠实的好朋友，萌萌大法，千秋万代，一统江湖。 责任编辑【威尔】
频谱仪原理及使用方法 频谱仪是一种将信号电压幅度随频率变化的规律予以显示的仪器。频谱仪在电磁兼容分析方面有着广泛的应用，它能够在扫描范围内精确地测量和显示各个频率上的信号特征，使我们能够“看到”电信号，从而为分析电信号带来方便。 1.频谱仪的原理 频谱仪是一台在一定频率范围内扫描接收的接收机，它的原理图如图1所示。 频谱分析仪采用频率扫描超外差的工作方式。混频器将天线上接收到的信号与本振产生的信号混频，当混频的频率等于中频时，这个信号可以通过中频放大器，被放大后，进行峰值检波。检波后的信号被视频放大器进行放大，然后显示出来。由于本振电路的振荡频率随着时间变化，因此频谱分析仪在不同的时间接收的频率是不同的。当本振振荡器的频率随着时间进行扫描时，屏幕上就显示出了被测信号在不同频率上的幅度，将不同频率上信号的幅度记录下来，就得到了被测信号的频谱。进行干扰分析时，根据这个频谱，就能够知道被测设备或空中电波是否有超过标准规定的干扰信号以及干扰信号的发射特征。 2.频谱分析仪的使用方法 要进行深入的干扰分析，必须熟练地操作频谱分析仪，关键是掌握各个参数的物理意义和设置要求。 (1)频率扫描范围 通过调整扫描频率范围，可以对所要研究的频率成分进行细致的观察。扫描频率范围越宽，则扫描一遍所需要时间越长，频谱上各点的测量精度越低，因此，在可能的情况下，尽量使用较小的频率范围。在设置这个参数时，可以通过设置扫描开始频率目”无“’。04朋和终止频率来确定，例如:startfrequeney=150MHz，stopfrequency=160MHz;也可以通过设置扫描中心频率和频率范围来确定，例如:eenterfrequeney=155MHz，span=10MHz。这两种设置的结果是一样的。Span越小，光标读出信号频率的精度就越高。一般扫描范围是根据被观测的信号频谱宽度或信道间隔来选择。如分析一个正弦波，则扫描范围应大于2f(f为调 制信号的频率)，若要观测有无二次谐波的调制边带，则应大于4f。 (2)中频分辨率带宽 频谱分析仪的中频带宽决定了仪器的选择性和扫描时间。调整分辨带宽可以达到两个目的，一个是提高仪器的选择性，以便对频率相距很近的两个信号进行区别，若有两个频率成分同时落在中放通频带内，则频谱仪不能区分两个频率成分，所以，中放通频带越窄，则频谱仪的选择性越好。另一个目的是提高仪器的灵敏度。因为任何电路都有热噪声，这些噪声会将微弱信号淹没，而使仪器无法观察微弱信号。噪声的幅度与仪器的通频带宽成正比，带宽越宽，则噪声越大。因此减小仪器的分辨带宽可以减小仪器本身的噪声，从而增强对微弱信号的检测能力。根据实际经验，在测量信号功率时，一般来说，分辨率带宽RBW宜为
正确使用频谱分析仪需注意的几点 首先，电源对于频谱分析仪来说是非常重要的，在给频谱分析仪加电之前，一定要确保电源接法正确，保证地线可靠接地。频谱仪配置的是三芯电源线，开机之前，必须将电源线插头插入标准的三相插座中，不要使用没有保护地的电源线，以防止可能造成的人身伤害。 其次，对信号进行精确测量前，开机后应预热三十分钟，当测试环境温度改变3—5度时，频谱仪应重新进行校准。 三，任何频谱仪在输入端口都有一个允许输入的最大安全功率，称为最大输入电平。如国产多功能频谱分析仪AV4032要求连续波输入信号的最大功率不能超过＋30dBmW（1W），且不允许直流输入。若输入信号值超出了频谱仪所允许的最大输入电平值，则会造成仪器损坏；对于不允许直流输入的频谱仪，若输入信号中含有直流成份，则也会对频谱仪造成损伤。 一般频谱仪的最大输入电平值通常在前面板靠近输入连接口的地方标出。如果频谱仪不允许信号中含有直流电压，当测量带有直流分量的信号时，应外接一个恰当数值的电容器用于隔直流。 当对所测信号的性质不太了解时，可采用以下的办法来保证频谱分析仪的安全使用：如果有RF功率计，可以用它来先测一下信号电平，如果没有功率计，则在信号电缆与频谱仪的输入端之间应接上一个一定量值的外部衰减器，频谱仪应选择最大的射频衰减和可能的最大基准电平，并且使用最宽的频率扫宽（SPAN），保证可能偏出屏幕的信号可以清晰看见。我们也可以使用示波器、电压表等仪器来检查DC及AC信号电平。 频谱分析仪的工作原理 频谱分析仪架构犹如时域用途的示波器,外观如图1.2所示,面板上布建许多功能控制按键,作为系统功能之调整与控制,系统主要的功能是在频域里显示输入信号的频谱特性.频谱分析仪
频谱仪的简单原理 通常我们要对即将传输或者已经接收的信号进行分析，以获取我们所需要的信息。为了获取不同的信息我们通常将信号放在不同的域进行分析，如下图所示： 对于时域（时间和幅值）的分析我们通常采用示波器，获取信号的幅度、周期、频率等信息；对于频域（频率和幅值）的分析我们通常采用频谱分析仪，获取信号的频率、功率、谐波、噪声等信息；剩下的则是时间和频率的域，我们称之为矢量域，我们可以通过适量分析仪获取信号的幅度误差、矢量误差、相位误差等信息。现在的许多频谱分析仪也兼有矢量分析仪的功能。 按照工作原理分，频谱有两种基本的类型：实时频谱仪和扫频调谐式频谱仪。实时频谱仪包括多通道滤波器（并联型）频谱仪和FFT频谱仪。扫频调谐式频谱仪包括扫描射频调谐型频谱仪和超外差式频谱仪。其中超外差频谱仪应用最为广泛。 1.超外差频谱仪
如下图所示是典型的超外差频谱分析仪的实现框图： 原始输入信号首先经过一个低通滤波器，随后经过衰减器到达混频器以后，与来自本振的信号相混频。因为混频器本身就是非线性器件，所以输出信号除了包含两个原始的信号之外，还包含谐波，以及原始信号与谐波的差信号与和信号。如果混频信号落在中频滤波器的通带范围内，则信号会被进一步处理，中频信号再经过放大、滤波后送到检波器检波．检波输出信号经视频滤波器滤波，成为与输入信号功率幅度相对应的视频信号，体现在显示屏的Y轴上；扫频控制器将扫描电压与本振频率对应起来，改变频谱仪本振频率的同时将改变显示屏X轴的扫描电压。这样，频谱仪就可以将输入信号在不同频率处的功率幅度大小体现在显示屏上了。 假设频率轴上有一个特定的窗口，那么只有进入到该窗口内的信号才能被检测到，这就是它的基本原理。如果窗口从频率点f1 扫描到频率点f2，就可以得到不同频率上的信号功率，也就得到了被测信号
---------------------------------------------------------------最新资料推荐------------------------------------------------------ 频谱仪原理与使用介绍 2008年4月频谱仪原理与使用介绍主讲： 李家杰2008年4月频谱测量的意义频谱仪的工作原理 频谱仪各主要组件的功能频谱仪的正确使用频谱仪的各项参数设置介绍频谱仪的校准利用频谱仪进行测量的一些技巧2008年4月频谱测量的意义频谱分析仪对于信号分析来说是不可少的。 它是利用频率域对信号进行分析、研究，同时也应用于诸多领域，如通讯发射机以及干扰信号的测量，频谱的监测，器件的特性分析等等，各行各业、各个部门对频谱分析仪应用的侧重点也不尽相同。 2008年4月频谱测量的意义科学发展到今天，我们可以用许多方法测量一个信号，不管它是什么信号。 通常所用的最基本的仪器是示波器---观察信号的波形、频率、幅度等，但信号的变化非常复杂，许多信息是用示波器检测不出来的，如果我们要恢复一个非正弦波信号F，从理论上来说，它是由频率F1、电压V1与频率为F2、电压为V2信号的矢量迭加。 2008年4月频谱测量的意义从分析手段来说，示波器横轴表示时间，纵轴为电压幅度，曲线是表示随时间变化的电压幅度，这是时域的测量方法。 如果要观察其频率的组成，要用频域法，其横坐标为频率，纵 1 / 24
轴为功率幅度。 这样，我们就可以看到在不同频率点上功率幅度的分布，就可以了解这两个（或是多个）信号的频谱分布。 Atf时域频域2008年4月频谱测量的意义所以说有了这些单个信号的频谱，我们就能把复杂信号再现、复制出来。 这一点在我们要对复杂信号进行频率测量和分析时，是非常重要的，是时域分析所无法实现的。 2008年4月频谱仪的工作原理从技术实现来说，目前有两种方法对信号频率进行分析。 其一是对信号进行时域的采集，然后对其进行傅里叶变换，将其转换成频域信号。 我们把这种方法叫作动态信号的分析方法。 特点是比较快，有较高的采样速率，较高的分辨率。 即使是两个信号间隔非常近，用傅立叶变换也可将它们分辨出来。 2008年4月频谱仪的工作原理但由于其分析是用数字采样，所能分析信号的最高频率受其采样速率的影响，限制了对高频的分析。 目前来说，最高的分析频率只是在10MHz或是几十MHz，也就是说其测量范围是从直流到几十MHz。 是矢量分析。 这种分析方法一般用于低频信号的分析，如声音，振动等。 2008年4月频谱仪的工作原理另一方法原理则不同。