50 years of transition-metal lasers: from ruby to Ti:sapphire
ICFO Colloquium ProgramICFO – The Institute of Photonic Sciences
Castelldefels (Barcelona), Spain July 5th, 2010Peter MoultonQ-Peak, Inc.
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
What makes things colored?
Electronic transitions giver rise to “colors” in the visible region of the electromagnetic spectrum
Electronic transition at red wavelength
Ener
gy ->
Absorbed energy by electrons– where does it go?
Ground state Excited state
Light Heat?
Light(fluorescence)
TimeEn
ergy Exp (-t / τ ), sometimes
Fluorescence quantum efficiency = Decay rate from light emission / Total decay rate
Stimulated emission (thanks to Einstein)
Excited state
Light
Stimulated emission
Stimulated emission competes with absorption
Loss
Gain
Thank you, 1917!
Physika Zeitschrift, Volume 18 (1917), pp 121-128
Pick a scheme and win a Nobel prize!
3-level laser
4-level laser
Also, for starters, find a system with high fluorescence quantum efficiencyand a narrow emission linewidth
What makes things colored? Part II
Organic
Inorganic
Chlorophyll – green coloring for leaves,from an organic molecule
Structure of chlorophyll a
Structure of methane
Transitions of 3d ions in solidsoften make inorganic colors
Number ofd electrons
Ion(s)
1 Ti3+
2 Ti2+, V3+
3 Cr3+, V2+
4 Cr2+, Mn3+
5 Fe3+, Mn2+
6 Fe2+, Co3+
7 Co2+, Ni3+
8 Ni2+
9 Cu2+
H
Li
Na
K Ca Sc
Rb Sr
Be
Mg
Ti V
Y
Cr Mn Fe Cu Zn Ga Ge
Zr Nb Mo
As
B C N
Al Si P
Tc Ru Cd In Sn SbRh Pd Ag
NiCo
Transition metals
H
Li
Na
K Ca Sc
Rb Sr
Be
Mg
Ti V
Y
Cr Mn Fe Cu Zn Ga Ge
Zr Nb Mo
As
B C N
Al Si P
Tc Ru Cd In Sn SbRh Pd Ag
NiCo
Transition metals
Sc [Ar] 3d14s2
Ti [Ar] 3d24s2
V [Ar] 3d34s2
Cr [Ar] 3d54s1
Mn [Ar] 3d54s2
Fe [Ar] 3d64s2
Co [Ar] 3d74s2
Ni [Ar] 3d84s2
Cu [Ar] 3d104s1
Zn [Ar] 3d104s2
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
d-electron orbitals – 5-fold degenerate in free space
Energy levels of ions with 3 d-shell electrons
d3 system fluorescence spectra
Ruby
Fluorescence lifetime vs. temperature for d3 systems
What is the quantum efficiency?
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
Early thoughts on ruby laser from Schawlow
• Interviewed by Joan Bromberg, 1984•
After we finished the paper, I knew that Townes and Cummins and later Abella and Heavens were going to work on trying to make a potassium optical maser at Columbia. And I never want to do what anybody else is doing, because I haven't much confidence in my ability to compete, and I don't like competing. And being at Bell Labs in the transistor era, you felt that if you could do anything in a gas, you could do it better in a solid. And so I started trying to learn about solids. And in fact, in that one paragraph in our paper that mentions that solids have broad bands for absorbing light and sharp lines to emit it, I had just learned that much; I knew that ruby was that way.
• Now, ruby was a common material around there because a lot of people were working on microwave masers. So you could go down the hall and find somebody who had a drawer full of rubies of various concentrations, and could borrow a few samples which you'd never return. So I just thought well, I'll get my feet wet, I'll try and learn something about this stuff, what's it all about. I had no idea of the theory, or anything at all about it. And I got hold of a copy of Pringsheim's book on Fluorescence and Phosphorescence. Which was one of these wonderful, thoroughly Germanic books that had all the references back to the early 1800s. It was very complete, but it didn't have the answers we wanted. At that time, I asked [lab director Al] Clogston if Icouldwork on that, and he said "Fine." Then later there was another incident in the fall of 1958 after — the fall of 1960, rather, after Maiman had published the pink ruby laser, I was thinking about the dark ruby, and I really knew quite a lot about it, and I knew that those satellite [dark ruby spectrum] lines, or "N" lines, were really very strong, stronger than the [pink ruby’s]"R" lines, and I just felt that that dark ruby maser that I had proposed really ought to work. So I asked Clogston if he thought I ought to try it out, and he said, "You owe it to yourself." So, we did, and it worked. Right away. And of course, I should have done it sooner.
Ruby quantum efficiency was thought by some to be low (Maiman disagreed)
First publication on laser
Stimulated Optical Radiation in RubyT. H. MAIMAN
Hughes Research Laboratories, A Division of Hughes Aircraft Co., Malibu, California.
Schawlow and Townes1 have proposed a technique for the generation of very monochromatic radiation in the infra-red optical region of the spectrum using an alkali vapour as the active medium. Javan2 and Sanders3 have discussed proposals involving electron-excited gaseous systems. In this laboratory an optical pumping technique has been successfully applied to a fluorescent solid resulting in the attainment of negative temperatures and stimulated optical emission at a wave-length of 6943 Å. ; the active material used was ruby (chromium in corundum).
1. Schawlow, A. L. , and Townes, C. H. , Phys. Rev., 112, 1940 (1958).2. Javan, A. , Phys. Rev. Letters, 3, 87 (1959).3. Sanders, J. H. , Phys. Rev. Letters, 3, 86 (1959).4. Maiman, T. H. , Phys. Rev. Letters, 4, 564 (1960).
Nature 187, 493 - 494 (06 August 1960)
From digital version of Nature article
Pictures of first ruby laser at Hughes
Bell Labs gets convinced it’s a laser
Hughes did more science
Sapphire (corundum, Al2O3) enabled ruby laser
CW ruby lasers with lamp pumping
Cryogenic cooling in 1962
Laser-pumped ruby laser
Ruby laser pumping Sm:CaF2
No comment
Legacy of early ruby laser development
• First laser• First Q-switched laser• First laser-driven nonlinear optics (harmonics, Raman, etc.)• First use of cryogenic cooling to improve thermo-optical and
spectral characteristics• First demonstration of laser pumping of a solid-state laser
– Argon-ion-pumped ruby laser– Ruby-laser-pumped Sm:CaF2 laser (first 5d-4f laser?)
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
Tunable lasers – organic dyes provided a start
Dye lasers, cw and pulsed
Isoelectronic traps in Te-doped CdS- try for a tunable laser, but Auger-process won
Rediscovery of first broadly tunable lasers,handicapped by cryogenic operation
Energy levels of divalent transition metals
Divalent Ni in MgF2 : Properties at 77 K
pump
Divalent Co in MgF2 :properties at 77 K
pump
Co:MgF2 boule and assorted TM-doped crystals grown at MIT Lincoln Laboratory
Photos of cryogenic lasers at MIT/LL (1978-1985)
Cryogenic operation of Co:MgF2 laser
First room-temperature operation from Co:MgF2
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
Bill Krupke suggested a possible material for a lamp-pumped fusion-driver laser – but no gain
Ce:YLF absorption/emission (1979)
(with Dan Ehrlich, Rick Osgood)
First Ce:YLF laser setup
One reviewer was skeptical
We did publish, and later made another laser
Excited-state absorption (ESA) a pervasive problem
Ce3+
Example of complexity in ESA calculations
Color-center laser levels inspired search for systems without ESA
Energy levels of single d electron in crystal
Number ofd electrons
Ion(s)
1 Ti3+
2 Ti2+, V3+
3 Cr3+, V2+
4 Cr2+, Mn3+
5 Fe3+, Mn2+
6 Fe2+, Co3+
7 Co2+, Ni3+
8 Ni2+
9 Cu2+
Early work on Ti in sapphire (1962)
MIT efforts studied defect diffusion using Ti
J. Am Ceramic Soc. 52, 331 (1969)
Ti:sapphire absorption/emission (1982)
400 500 600 700 800 900 1,0000
0.2
0.4
0.6
0.8
1
0
0.2
0.4
0.6
0.8
1
WAVELENGTH (nm)
ABSO
RPT
ION
CO
EFFI
CIE
NT
(arb
. uni
ts)
FLU
OR
ESC
ENC
E IN
TESI
TY (a
rb. u
nits
)
Fluorescencelifetime3.2 usec
Jahn-Teller splitting for upper and lower levelsleads to broadened transitions
First Ti:sapphire laser operation
Ti:sapphire - early photos in 1982-3
MIT couldn’t afford (!) to patent Ti:sapphire
Parasitic absorption was a party spoiler
400 600 800 1,000 1,2000
1E-20
2E-20
3E-20
4E-20
5E-20
6E-20
7E-20
WAVELENGTH (nm)
CR
OSS
SEC
TIO
N (c
m^2
)
AB
S. C
OEF
FIC
IEN
T (a
rb. u
nits
)
PI
SIGMA
Work at LL examined Ti3+-Ti4+ as culprit
Predictions that were right
MIT LL Solid State Research 1982:3
Predictions that were (mostly) wrong
Technology genealogy
V:MgF2FUSION DRIVER
COLOR-CENTERLASERS
Ti:SAPPHIRELASER
Cr:LiSAFLASER
ESA-crippled
Simple energy levels
Aha!
Understand ESA
Try again
Not a good fusion driver, but...
Crystal engineering?
Livermore
Bell
Lincoln
Livermore
Outline
• Color and light• Quick review of transition-metal spectroscopy• The ruby laser and its consequences• Divalent transition-metal lasers• Ti:sapphire background• Impact of Ti:sapphire lasers
200-W average power from lamp-pumped Ti:sapphire
My own group’s work on Ti:sapphire
Laser-pumped, high-energy, ns-pulse Ti:sapphire laser
Developed with NASA Langley, DARPA support, 1986-1992
0
50
100
150
200
250
300
350
400
450
500
0 200 400 600 800 1000 1200 1400
Green pump energy (mJ)
Ti:s
apph
ire
outp
ut e
nerg
y (m
J)
790 nm727 nm911 nm960 nm
10-20-ns pulse duration
diffraction-limited
Pump #1
Pump #2GRM
Ti:sapphirecrystals
Prisms
Diodeseed
Pump #1
GRM
Ti:sapphirecrystals
Prisms
DiodeseedSeed
Tuning curve of Titan-CW laser pumped by argon-ion laser
700 750 800 850 900 950 1,000 1,050 1,100 1,1500
0.5
1
1.5
2
Wavelength (nm)
Pow
er O
utpu
t (W
)
7 W pump power
Rare-earth levels and Ti:sapphire tuning
Key tool in development of Er:fiber amplifiers
Ti:sapphire gain bandwidth support 5 fs pulses
600 700 800 900 10000
0.2
0.4
0.6
0.8
1
WAVELENGTH (nm)
INTE
NSI
TY (a
rb. u
nits
)
GAINPI
SIGMA
98 THz (4.4 fs)
Kerr-lens modelocking (KLM) providesa fast switch to enable fs-pulse modelocking
Ti:sapphire ultrafast lasersreplaced dye lasers in the 90’s
Counting optical cycles
Significance of femtosecond lasers
"for his studies of the transition states of
chemical reactions using femtosecond spectroscopy"
The Nobel Prize in Chemistry 1999
Ahmed H. ZewailEgypt and USA
California Institute of Technology (Caltech)
Pasadena, CA, USA
b. 1946
Time Domain ↔ Frequency Domain
2πδ= Δφ frep
I(f)
f
δ
0
frepI(f)
f
δ
0
frep
•Frequency modes of the fs pulse are offset from fn=0=0 by δ
Frequency Domain
TimeDomain
2Δφ
t
E(t)
• How can we control the absolute frequencies (and hence the group-phase velocities)? Self-referencing
Frequency
FundamentalSpectrum
SecondHarmonic Spectrum
m Δν+δn Δν+δ
2(m Δν+δ)Δν
460 480 500 520 540
Fundamental-Second Harmonic
Beats
Repetition Rate
RF
Pow
er (1
0 dB
/div
)Frequency (MHz)
D. J. Jones et al, Science 288 p 635 28 April 2000
J. Reichert et al., Opt. Comm. 172 pp 59–68 15 Dec 1999
H. Telle et al., Appl. Phys. B 69, 327–332 8 Sept 1999
Locking via Self-ReferencingTechnique
Beat frequency at overlap = δ
StockholmDecember 10, 2005
Hansch and Hall win Nobel Prize for Optical Combs
Chirped pulse amplification (CPA)
Courtesy: Wikipedia1985 (G.Mourou & D.Strikland)
Under the hood of ahigh-power Ti:sapphire CPA system
Size does matter for high-energy systems
Photograph of Ti:sapphire-generated filament for lidar
Attosecond pulses, high-harmonic generation
CPA pushes to a Zettawatt (courtesy Mourou)
Ti:sapphire laser - highlights
• Broadly tunable (650-1100 nm) output used widely for scientific and applied linear and nonlinear spectroscopy of gases and condensed media, atmospheric research
• Mode-locked output <10 fs has probed ultrafast dynamics of media (Zewailawarded Nobel Prize in Chemistry for work on molecules)
• Mode-locked systems also can generate new optical frequency standardsand allow measurement accuracies of a part in 1018
• Amplified mode-locked lasers (with CPA) have approached Petawatt (1015
W) of output (30 J in 30 fs) to study laser-matter interactions at extremely high intensities, generate x-rays
• Commercial laser sales are on the order of 6000 systems, about $500 million (update, approaching $1B).
Thank you!