In-vacuum scattered light reduction with cupric oxide surfaces for sensitive fluorescence detection

1.2 MB·

by  Eric B. Norrgard

In-vacuum scattered light reduction with cupric oxide surfaces for sensitive
fluorescence detection
E. B. Norrgard,∗ N. Sitaraman,† J. F. Barry,‡ D. J. McCarron, M. H. Steinecker, and D. DeMille
Department of Physics, Yale University, P.O. Box 208120, New Haven, Connecticut 06520, USA
We demonstrate a simple and easy method for producinglow-reflectivity surfaces that are ultra-
highvacuumcompatible,maybebakedtohightemperatures,andareeasilyappliedevenoncomplex
surface geometries. Black cupric oxide (CuO) surfaces are chemically grown in minutes on any
coppersurface,allowingforlow-cost,rapidprototypingandproduction. Thereflectivepropertiesare
measuredtobecomparabletocommerciallyavailableproductsforcreatingopticallyblacksurfaces.
6
1 We describe a vacuum apparatus which uses multiple blackened copper surfaces for sensitive, low-
0 background detection of molecules using laser-induced fluorescence.
2

n INTRODUTION undesired reflective surfaces may be covered with black-
a
enedcopperfoilsorsheetsformedintoanydesiredshape.
J
0 Laser-induced fluorescence (LIF) detection is a sim- BlackenedcopperissuitableinUHVfromcryogenictem-
2 ple yet powerful technique frequently used in gas-phase peratures up to 500◦C [14].
atomic and molecular spectroscopy [1–4]. In many in-
]
et teresting situations, LIF signals are fairly small. For ex- BLACKENING METHOD
ample, in the gas phase, rovibrational branching from
d
- molecularelectronicstatesmayleadtoLIFsignalsof.1
s photon per molecule. Modern experiments with atoms Copper has two common oxidization states: the red
n
[5], molecules [6], and ions [7] frequently aim to detect Cu2O (cuprous oxide) and the black CuO (cupric ox-
i
. ide). The black CuO is effectively grown on the surface
s small numbers of particles—sometimes as few as one.
c In such cases, scattered laser light (and the noise from of metallic copper by immersing the metal in a solution
si it) can seriously degrade the signal-to-background and composedof100gofNaOHand100gofNaClO2perliter
y of deionized water [15]. The final step of the reaction,
signal-to-noise ratios in LIF detection.
h
p Blackenedsurfacesareoftenusedtosuppressscattered Cu(OH)2(s)−−−→CuO(s)+H2O(l), (1)
[ lightin LIFdetectionexperiments [2,3,8]. Recently, we heat

haveinvestigatedopticallyblackmaterialsforstraylight
1 growsthe black CuO [16]. The solution’s temperature is
v suppression to facilitate efficient detection of molecules maintained between 95◦C and 100◦C during the black-
8 viaLIF(foruseinexperimentsonlaserslowing[9],cool- ening process. Thermal and concentration gradients are
0 ing [10], and trapping [11–13] of SrF molecules). Here,
minimizedbyusingamagneticstirbarinthe blackening
3 we present a method for producing ultra-high vacuum
solution. Immersing the beaker holding the blackening
5
(UHV)compatiblelow-reflectivitysurfacesthatissimple
0 solution in a heated outer water bath also reduces ther-
toimplementandveryflexibleinitsutility. Ourmethod,
. malgradients. The outer bathis saturatedwithNaClto
1 which we refer to as “copper blackening”, uses a simple allow its temperature to exceed 100◦C without boiling
0
chemicaltreatmenttoproduceblack,dendriticcupricox-
6 and potentially contaminating the solution.
ide(CuO)onthesurfacesofanypre-formedcopperpart.
1 The material used for all parts described in this pa-
v: Wedescribethemethodandpresentmeasurementsofthe per is oxygen-freehighconductivity (OFHC) copper (al-
reflectance properties of these prepared surfaces. Mea-
i loy 101), selected for its desirable UHV properties. As
X surementsonblackenedcopper surfacesarecomparedto
described below, we tested OFHC copper prepared with
r measurementsonwell-characterizedcommercialsurfaces.
surfacesbothas-finished,andsandblastedtoincreasethe
a
Blackened copper has a number of attractive proper-
roughness. We have also performed the blackening pro-
tiesforstraylightsuppressionsolutions. Copperiseasily
cess on alloy 110 Cu and on brass parts, which might be
machined or formed, and the surfaces may be grown in
usefulinapplicationswhereUHVisnotaconcern(while
minutes on parts of any manufactured geometry. Vac-
we have not measuredthe reflectance properties of these
uum componentsmaybe made ofblackenedcopper,and
parts,theyappearsimilarlyblackbyeyetothe101alloy
measured in this work). Most steels and titanium are
minimallycorrodedbythesolution[17]andmaybeused
toholdorremovepartsfromthesolution,butaluminum
∗ Electronicaddress: [email protected]
rapidly dissolves in NaOH.
† Currentaddress: DepartmentofPhysics,CornellUniversity,142
SciencesDr.,Ithaca, NewYork14853,USA Prior to blackening, parts are cleaned in an ultrasonic
‡ Currentaddress: Harvard-SmithsonianCenterforAstrophysics, bath of ∼1% Citranox in deionized water for one hour
60GardenStreet, Cambridge,Massachusetts 02138, USA. toremoveanysurfacecontamination,followedbyarinse
2
in a deionized water bath. Parts are then immediately is the distance from the scattering surface to the detec-
placed in the blackeningsolution for 10 minutes (by eye, tor, θi and θr are the incident and reflected polar angles,
parts blackened for 10 minutes are not distinguishable respectively, and φr is the relative reflected azimuthal
from parts blackened for 5 or 15 minutes; under similar angle.
conditions,theoxidelayerthicknesswasobservedtosat- If the geometric collection efficiency for the scattering
uratein2.5minutesinRef.[18]). Partsarethenremoved source is large (i.e. the detector subtends a large solid
fromthe solution,rinsedinafreshdeionizedwaterbath, angle), or if there are multiple scattering surfaces at dif-
and placed in an ultrasonic acetone bath for one hour. ferentanglesrelativetotheincidentlight,thedirectional-
After a final fresh deionized water rinse, parts are dried hemispherical reflectance (DHR) ρ(θi), defined as
with a gentle stream of dry N2.
CuO is knownto grow on surfaces in a microstructure ρ(θi)= fr(θi,θr,φr)cos(θr)dΩr, (3)
Z
of fine dendrites by a number of wet-chemical [16] and
electrochemical methods [19, 20]. These dendrites effec- is a more useful characteristic of the surface.
tivelytrapincidentlight[21],furtherreducingreflection. For a simple relative comparison of materials, we di-
Care must be taken when handling the blackened parts, rectalaserbeamwith1/e2 diameter≈1mmtointersect
as the CuO dendrites are easily crushed by the pressure the axisofrotationofaturntable,withthe sample ofin-
of touching the surface, rinsing with a squirt bottle, or terestatits center. Thesampleanddetectorphotodiode
dryingwithahighpressureofcompressedgas. Crushing may be rotated independently about this axis. A chop-
thedendritesisobservedtoresultinalocallylessabsorp- perwheelmodulatestheincidentlightpowerPi at3kHz,
tive surface. However, the dendrites are not observed to andthereflectedpowerPr isextractedfromthe detector
flake from the surface when crushed, or when subjected photodiodesignalusingalock-inamplifier. Afractionof
to mechanical or ultrasonic vibration. the beam Pm is sentto a monitor photodiode to account
In some instances, having a partwith only some areas for source intensity fluctuations. Prior to a reflectance
blackenedis desirable. While wefind the thinCuO layer measurement,thesignalandmonitorphotodiodesrecord
may be easily removed by mechanical means, such as Pi andPm,respectively. Duringthereflectancemeasure-
sandingormachining,itisnotalwayspossibletoperform ment, Pi isnormalizedtothe monitorphotodiodesignal.
suchoperationswithoutcrushingthedendritesinthear- The simplicity ofthis setup limits us to measurements
eas where blackening is desired. We have investigated with φr=0◦ (forward scattering) or φr=180◦ (back
techniquesformaskingareastoinhibitCuOgrowth,and scattering). To describe these data, we define the
found adhesive polytetrafluoroethylene (PTFE) tape to bidirectionalazimuth-independentreflectiondistribution
provide good results (see below). PTFE does not react
with the blackening solution, and the acrylic adhesive
leaveslittle visible residue; whatdoes remainis removed
z
intheultrasonicacetonebathtypicallyusedinourclean-
A
ingprocedure. Theblackeningsolutioncreepsonlyafew
mmundertheedgesofthetapeduringa10minuteblack- θ
ening, leaving a reasonably sharp boundary. i θ D
r
METHOD FOR MEASURING REFLECTANCE
When attempting to eliminate background light, the
relative positions of the light source, scattering surface,
anddetectorcansignificantlyaffecttherelativeintensity
of scattered light. For example, surfaces typically have
dramatically different reflectance properties near normal
incidence and near grazing incidence. Because of this,
optical reflectance is typically described in terms of the
bidirectional reflectance distribution function (BRDF)
fr(θi,θr,φr), defined as
PrD2
fr(θi,θr,φr)= . (2) FIG. 1. (Color online) Coordinate system used in measure-
PiAcosθr
ments. Thez-axisisthesurfacenormal. Incidentlighttravels
Here, Pr is the power reflected to the detector from inci- in the x−z plane at angle θi to the normal. Reflected light
dentpowerPi. Thegeometricalquantitiesusedtodefine is measured at polar angle θr and azimuthal angle φr on a
fr aredepicted inFig.1: Ais the areaofthe detector, D detector of area A a distance D away.
3
λ = 532 nm λ = 1064 nm
0.1
Blackened Stock Cu
-1sr) Blackened Sandblasted Cu
F ( Bright White 98
D Acktar Metal Velvet
R Black Felt
AI
B0.01
1E-3
-80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
θ (degrees) θ (degrees)
r r
FIG. 2. (Color online) Bidirectional azimuth-independent reflection distribution function vs reflection angle for materials
investigatedwithλ=532nm(left)andλ=1064nm(right)atnormalincidence(θi=0◦). ThedashedlinemarksaLambertian
surface with unit reflectivity.
θ = 0 degrees θ = 70 degrees
i i
Sand-
Stock blasted
0.1 405 nm
532 nm
650 nm
-1sr) 1064 nm
F (
D
R
AI0.01
B
1E-3
-100 -80 -60 -40 -20 0 20 40 60 80 -80 -60 -40 -20 0 20 40 60 80
θ (degrees) θ (degrees)
r r
FIG.3. (Coloronline)Bidirectionalazimuth-independentreflectiondistributionfunctionvsreflectionangleforblackenedstock
copper (filled markers) and blackenedsandblasted copper (open markers) at normal incidence (left) and 70◦ incidence (right).
Material λ =532nm λ =1064nm λ Stock SandBlasted
Bright White98 0.95 0.95 405nm 0.015 0.012
Blackened Cu, Stock 0.008 0.07 532nm 0.008 0.008
Blackened Cu, Sandblasted 0.008 0.06 650nm 0.010 0.013
Black Felt 0.006 0.17 1064nm 0.07 0.06
AcktarMetal Velvet 0.005 0.024
TABLEII.Directional-hemispherical reflectionρ(θi =0◦)for
TABLE I. Directional-hemispherical reflection ρ(θi =0◦) for blackened stock copper and blackened sandblasted copper vs
materials investigated. wavelength λ.
4
function (BAIRDF) as Fr(θi,θr)=fr(θi,θr,φr), with θr ing would tend to randomizethe surfacenormal, andin-
positive (negative) for φr=0◦ (φr=180◦). To assign deed no specular reflection component was observed for
a value for the DHR from the BAIRDF at normal blackened sandblasted copper for any angle of incidence
incidence (θi=0◦), we perform the integral over φr in we were able to test, θi < 89◦. However, the removed
Eq.3 by averaging over the measured values for φr=0◦ specular reflection is accompanied by increased forward
and φr=180◦ (This is essentiallyequivalent to assuming (positive θr) diffuse scatter. For θi = 70◦ (an incident
azimuthally symmetric scattering for the case of normal angle just below the onset of significant specular reflec-
incidence). tion in blackened stock copper), the stock finish has less
forward scatter than the sandblasted copper, while back
scatter is lower for the sandblasted copper.
DISCUSSION OF RESULTS Fig.4a shows stock and sandblasted copper samples
beforeandafterblackeningfortestsinthiswork. Copper
In Fig.2, we compare Fr at normal incidence for five masked by adhesive PTFE tape is shown in Fig.4e. Re-
materials at wavelengths λ=532nm and 1064nm. Val- cently, we have used blackened copper surfaces (Fig.4b-
ues for ρ(θi=0◦) are tabulated in Table I. To confirm d) to reduce background scattered light in experiments
that the calculated Fr and ρ values were reliable, ρ was imagingmagneto-opticallytrappedSrFmoleculesbyLIF
measured for a near-Lambertian, high-reflectivity sur- detection [11–13]. A schematic of the detection region
face (Bright White 98). We measured ρ(θi=0◦)=0.95 used in Ref.[13] is shown in Fig.5. A 14mm 1/e2 di-
at532nm, comparedtothe manufacturer-specifiedvalue ameter laser beam (the beam is clipped by 23mm diam-
0.96 typical at 550nm (no test data at 1064nm were eter optics external to the vacuum chamber) containing
available from the manufacturer). For a commer- 110mWofλ=663nmlightispassed6timesthroughthe
cial light-absorbing blackened surface provided on alu- vacuum chamber (once eachalong three orthogonalaxes
minum foil (Acktar Metal Velvet, AMV), we measure and then retroreflected). Excitation by the ≈660mW of
ρ(θi=0◦)=0.005 at 532nm, also in fair agreement with λ=663nmlightinthecenterofthechamberinducesflu-
the manufacturer’s specified typical value of 0.006. orescence from the trapped molecules at the same wave-
Inallcasesinvestigatedatnormalincidence,theAMV length. Light is collected by a 2”-diameter, 150-mm-
provided the blackest surface. However, the blackened focal-length spherical singlet lens placed directly outside
copper had only marginally higher reflectance than the the vacuum chamber, followed by a 50-mm-focal-length,
AMV, and proved generally comparable to standard f/0.95cameralens. Themagnificationfactorattheimag-
black felt cloth. At normal incidence, no significant dif-
ferencewasmeasuredbetweenblackenedcopperstarting
with a stock finish and copper with its surface rough-
ened by sandblasted prior to cleaning and blackening.
Atλ=532nm,blackenedcopperis≈50%morereflective
thanblackfelt,and≈75%morereflectivethanAMV.At
1064nm, blackened copper is ≈2× less reflective than
black felt, and ≈3× more reflective than AMV.
In Fig.3, we plot the measured BAIRDF as a func-
tion of reflected angle θr for incident angle θi=0◦ (left)
and θi=70◦ (right) for four common laser wavelengths
λ: 405, 532, 650, and 1064 nm. The DHR for normal
incidence is given in Table II. As a consistency check of
ourBAIRDFdata,wenotethatbyreciprocityweexpect
Fr(θi,θr) = Fr(θr,θi). ForallcasesinFig.3,thesevalues
agree to within 20%.
At normal incidence, the blackened stock and sand-
blasted copper display reflectance nearly independent of
θr. Both surface preparationmethods produce compara-
ble values of ρ(θi = 0◦) for all wavelengths tested. FIG. 4. (Color online) Examples of blackened copper parts.
(a)Stock(left)andsandblasted(right)coppersamplesbefore
Blackened copper produced from stock material was
(top) and after (bottom) the blackening process. (b) CF full
found to specularly reflect for θi &75◦, making it highly nipples with blackened copper inserts. (c) Blackened copper
ineffective at reducing scattered light at near-grazingin- diskcoversCFflangewithholeinthecenterforopticalaccess.
cidence. Sandblasting (using sand with typical parti- (d) Blackened copper rods serves as heat (thick outer rods)
cle diameter ∼10µm) produces a diffuse surface over andelectronicsconduits(thininnerrods)forin-vacuummag-
a much larger scale than the typical scale of the den- net circuit boards. (e) Copper sheet with midsection masked
drites (∼100nm [16, 21]). We expected that sandblast- from the blackeningprocess using adhesivePTFE tape.
5
thisdiskwasreplacedwiththeassemblyshowninFig.4d,
whereblackenedcopperrodsserveasheatconduits(thick
outerrods)forin-vacuumhigh-powercircuitboards. We
found that the unmodified blackened surface does not
allow for good thermal contact. However, deliberately
crushing the CuO dendrites on the thermally connected
surface and applying a 0.004”-thick indium thermal in-
terface layer (TIL) gave nearly the same thermal con-
es ductivity as obtained with a typical method for making
heat links with unblackened copper (sanding the copper
surface with 2000 grit sandpaper and applying an in-
dium TIL). Using an indium TIL without crushing the
dendrites gave reduced conductivity (roughly 2× less),
similartothatofsandedcopperwithoutindiumpresent.
During normal trap operation, we measure the
maximum photon scattering rate per molecule to be
Rsc=2.5×106s−1. Analysis is restricted to a region of
interest of typically 50×50 pixels (3.2×3.2mm) where
the majority of the LIF signal is imaged. The back-
ground scattered light signal is typically B = 1.5×104
e−/pixel/s. For a typical te=60ms exposure, the noise
FIG.5. (Coloronline)Schematicofvacuumchamberusedto
isusuallylimitedbyshotnoiseinthisbackground,atthe
detect SrF molecules in a magneto-optical trap using laser-
inducedfluorescence. SeveralblackenedCu surfaces are used levelNB ≈30e−/pixel. Undertheseconditions,thelaser
to reduce scattered light. Blackened Cu heat links and elec- intensity noise,camera readnoise, andcamera dark cur-
4
tronics conduits are placed between high-power in-vacuum rent are approximately 4, 8 and 10 times smaller than
circuit boards. The thickness of blackened surfaces are exag- NB, respectively. Integrating over the region of interest
gerated for clarity. gives a signal-to-noise ratio (SNR) of 0.39 per molecule
for a single trap loading shot. With this system, we
haveobservedMOTsofasfew as 17molecules[13], with
ing position is M=0.45. Directly opposite the camera SNR≈110 when averagingover 300 successive shots.
is a blackened Cu screen against which the molecules
are imaged. The total light detection efficiency for our
system (including geometry, optical losses, and camera CONCLUSION AND OUTLOOK
quantum efficiency) is 0.4% [11].
Scattering from the viewports is minimized by using We have used a procedure to produce blackened cop-
highopticalquality(10-5scratchdig)glasswithananti- per surfaces [15], suitable for use in UHV applications.
reflection V-coating at 663nm. Standard vacuum nip- We measure these surfaces to be only a few times
ples with blackenedcopper sheet inserts (Fig.4b) reduce more reflective than commercial UHV light-absorbing
stray light in two ways. First, they reduce the power foil. While such foils are available with low-outgassing
of light scattered directly from viewport surface imper- adhesives,theytypicallyhavestricttemperaturerequire-
fections into the LIF detection region, simply by placing ments (40◦C.T .150◦C). Blackened copper requires
the viewports further from the imagingchamber (≈10”) no adhesive, can be baked at high temperature in vac-
than they would be in the absence of the nipple (≈5”). uum, and can be applied chemically to complex surfaces
Second,theblackenedinsertsincreasethelikelihoodthat as well as to flexible foils and sheets. Sandblasting prior
light incident on the walls of the nipple (either directly to blackening is found to decrease diffuse back scatter,
or through scattering) is absorbed before reaching the increase diffuse forwardscatter, and remove specular re-
imagingregion. Backgroundlightisadditionallyreduced flection.
by baffles [2] formed by blackenedCu rings with a knife- The advantages of vacuum-compatibility, rapid proto-
edgedinnerdiametermachinedtobe26mm,whichsitin- typing, and low cost make this technique ideally suited
side the imaging chamber flush with the vacuum-sealing for scattered light suppression in experiments with LIF
Cu gasket. signalsfromagas-phasesource. Recently,blackenedcop-
Fig.4cshowsablackenedcopperdiskforstraylightab- per surfaces installed in our vacuum chamber reduced
sorption (not shown in Fig.5) which covered the entire stray light noise to a level sufficient to detect a small
bottom of the molecule imaging chamber in [11, 12]. A handful of molecules via LIF [13]. We also plan to use
holeinthe centerprovidesopticalaccessforthe twover- this technique to reduce scattered light background in
ticalpassesofthe663nmlight. FortheworkinRef.[13], various types of precision measurements using LIF from
6
molecular beams [22–24], and we expect this technology Y. V. Gurevich, P. W. Hess, N. R. Hutzler, E. Kirilov,
will be useful in similar experiments. I.Kozyryev,B.R.O’Leary,C.D.Panda,M.F.Parsons,
E. S. Petrik, B. Spaun, A. C. Vutha, and A. D. West,
The authors thank E.R.Edwardsforcontributions to- 343, 269 (2014).
ward the construction of the vacuum chamber. The [23] S. B. Cahn, J. Ammon, E. Kirilov, Y. V. Gurevich,
authors acknowledge financial support from ARO and D.Murphree,R.Paolino,D.A.Rahmlow,M.G.Kozlov,
and D.DeMille, Phys.Rev.Lett. 112, 163002 (2014).
ARO (MURI). EBN acknowledges funding from the
[24] L. R. Hunter, S. K. Peck, A. S. Greenspon, S. S. Alam,
NSFGRFP.NSacknowledgesfunding fromYaleCollege and D.DeMille, Phys.Rev.A 85, 012511 (2012).
Dean’s Research Fellowship.
[1] J. L. Kinsey, Annual Review of Physical Chemistry 28,
349 (1977).
[2] G.Scoles,AtomicandMolecularBeam Methods (Oxford
UniversityPress, New York,1988).
[3] H. Pauly, Atom, Molecule, and Cluster Beams I: Basic
Theory, Production, and Detection of Thermal Energy
Beams (Springer-Verlag, 2000).
[4] W.Demtr¨oder,LaserSpectroscopy2: ExperimentalTech-
niques, 5th ed. (Springer-Verlag, 2015).
[5] C.GrossandI.Bloch,“Microscopy ofmany-bodystates
inopticallattices,”inAnnual Review ofColdAtomsand
Molecules,Vol.3(WorldScientificPublishingCo.,2015).
[6] W. E. Moerner and D. P. Fromm, Review of Scientific
Instruments74, 3597 (2003).
[7] D. Leibfried, R. Blatt, C. Monroe, and D. Wineland,
Rev.Mod. Phys. 75, 281 (2003).
[8] G. Hall, K. Liu, M. J. McAuliffe, C. F. Giese, and
W.R.Gentry,TheJournalofChemicalPhysics81,5577
(1984).
[9] J. F. Barry, E. S. Shuman, E. B. Norrgard, and D. De-
Mille, Phys. Rev.Lett. 108, 103002 (2012).
[10] E.S.Shuman,J.F.Barry, andD.DeMille,Nature467,
820 (2010).
[11] J.F.Barry,D.J.McCarron,E.N.Norrgard,M.H.Stei-
necker, and D. DeMille, Nature512, 286 (2014).
[12] D.J. McCarron, E. B. Norrgard, M. H.Steinecker, and
D.DeMille, NewJournal of Physics 17, 035014 (2015).
[13] E.B.Norrgard,D.J.McCarron,M.H.Steinecker,M.R.
Tarbutt, and D. DeMille, (2015), arXiv:1511.00930.
[14] P.D.KirschandJ.G.Ekerdt,JournalofAppliedPhysics
90, 4256 (2001).
[15] W.Meyer,“Processforblackeningcopperorcopperalloy
surfaces,” (1944), US Patent 2,364,993.
[16] G. Filipiˇc and U. Cvelbar, Nanotechnology 23, 194001
(2012).
[17] Sandvik Materials Technology, “Sodium hy-
droxide corrosion tables,” Available online at
http://smt.sandvik.com/en/materials-center/corrosion-
tables/sodium-hydroxide/ (viewed 12 Jan 2016).
[18] M.Lebbai,J.-K.Kim,W.Szeto,M.Yuen, andP.Tong,
Journal of Electronic Materials 32, 558 (2003).
[19] W.ShaoandG.Zangari,TheJournalofPhysicalChem-
istry C 113, 10097 (2009).
[20] Y. Hu, X. Huang, K. Want, J. Liu, J. Jiang, R. Ding,
X.Ji, and X.Li, Journal ofSolid StateChemistry 183,
662 (2010).
[21] Y.Xia,X.Pu,J.Liu,J.Liang,P.Liu,X.Li, andX.Yu,
J. Mater. Chem. A 2, 6796 (2014).
[22] The ACME Collaboration, Baron, J., J. Baron, W. C.
Campbell, D. DeMille, J. M. Doyle, G. Gabrielse,