# Characterizing a fiber-based frequency comb with electro-optic modulator.

**ABSTRACT** We report on the characterization of a commercial- core fiber-based frequency comb equipped with an intracavity free-space electro-optic modulator (EOM). We investigate the relationship between the noise of the pump diode and the laser relative intensity noise (RIN) and demonstrate the use of a low-noise current supply to substantially reduce the laser RIN. By measuring several critical transfer functions, we evaluate the potential of the EOM for comb repetition rate stabilization. We also evaluate the coupling to other relevant parameters of the comb. From these measurements, we infer the capabilities of the femtosecond laser comb to generate very-low-phase-noise microwave signals when phase-locked to a high-spectral-purity ultra-stable laser.

**0**Bookmarks

**·**

**104**Views

- Anthony Bercy, Saïda Guellati-Khélifa, Fabio Stefani, Giorgio Santarelli, Christian Chardonnet, Paul-Eric Pottie, Olivier Lopez, Anne Amy-Klein[Show abstract] [Hide abstract]

**ABSTRACT:**We demonstrate in-line extraction of an ultra-stable frequency signal over an optical link of 92-km of installed telecommunication fibers, following the proposition of G. Grosche in 2010. We show that the residual frequency noise at the extraction end is noticeably below that at the main link output when the extraction is near the input end, as expected from a simple model of the noise compensation. We obtain relative frequency instabilities, expressed as overlapping Allan deviation, of 8x10-16 at 1 s averaging time and a few 10-19 at 1 day. These results are at the state-of-the-art for a link using urban telecommunication fibers. We also propose an improved scheme which delivers an ultra-stable signal of higher power, in order to feed a secondary link. In-line extraction opens the way to a broad distribution of an ultra-stable frequency reference, enabling a wide range of applications beyond metrology.Journal of the Optical Society of America B 02/2014; 31(4). · 2.21 Impact Factor - SourceAvailable from: Bérengère ArgenceBruno Chanteau, Olivier Lopez, Wei Zhang, Daniele Nicolodi, Bérengère Argence, Frédéric Auguste, Michel Abgrall, Christian Chardonnet, Giorgio Santarelli, Benoît Darquié, Yann Le Coq, Anne Amy-Klein[Show abstract] [Hide abstract]

**ABSTRACT:**We present a new method for accurate mid-infrared frequency measurements and stabilization to a near-infrared ultra-stable frequency reference, transmitted with a long-distance fibre link and continuously monitored against state-of-the-art atomic fountain clocks. As a first application, we measure the frequency of an OsO4 rovibrational molecular line around 10 $\mu$m with a state-of-the-art uncertainty of 8x10-13. We also demonstrate the frequency stabilization of a mid-infrared laser with fractional stability better than 4x10-14 at 1 s averaging time and a line-width below 17 Hz. This new stabilization scheme gives us the ability to transfer frequency stability in the range of 10-15 or even better, currently accessible in the near-infrared or in the visible, to mid-infrared lasers in a wide frequency range.New Journal of Physics 04/2013; 15(7). · 4.06 Impact Factor -
##### Article: Dual photo-detector system for low phase noise microwave generation with femtosecond lasers.

[Show abstract] [Hide abstract]

**ABSTRACT:**Low phase noise microwave signals can be generated by photo-detecting the pulse train of an optical frequency comb locked to a high spectral purity continuous-wave optical reference. Amplitude-to-phase noise conversion is, however, a well-known limitation to this technique. Great care is usually required to overcome this constraint due to its strong dependence on the impinging optical power. Here we demonstrate the combined use of "magic point" operating conditions of photodetectors, pulse repetition rate multipliers, and coherent addition of microwave signals to realize a microwave extraction device largely immune to amplitude-to-phase conversion effects over a large range of impinging optical powers.Optics Letters 03/2014; 39(5):1204-7. · 3.39 Impact Factor

Page 1

1

Characterizing a fiber-based frequency comb with

electro-optic modulator.

W. Zhang1, M. Lours1, M. Fischer2, R. Holzwarth2, G. Santarelli1 and Y. Le Coq1

1LNE-SYRTE, Observatoire de Paris, CNRS, UPMC, 61 avenue de l'Observatoire, Paris,

France

2MenloSystems GmbH, Am Klopferspitz 19a, D-82152 Martinsried, , Germany

Abstract

We report on the characterization of a commercial-core fiber-based frequency comb equipped

with an intracavity electro-optic modulator (EOM). We investigate the relationship between

the noise of the pump diode and the laser relative intensity noise (RIN) and demonstrate the

use of a low noise current supply to substantially reduce the laser RIN. By measuring several

critical transfer functions, we evaluate the potential of the EOM for comb repetition rate

stabilization. We also evaluate the coupling to other relevant parameters of the comb. From

these measurements we infer the capabilities of the femtosecond laser comb to generate very

low phase noise microwave signals when phase locked to a high spectral purity ultra-stable

laser.

Page 2

2

1.Introduction.

Femtosecond lasers have revolutionized the field of time and frequency metrology by

providing a phase coherent link between a large span of optical and microwave frequencies

[1-7]. These so called optical frequency combs are widely used for optical frequency

measurements with uncertainty close to the limitations of the caesium fountain clocks [8-11],

as well as frequency comparisons between different optical frequency standards [12,13]. By

transferring the spectral purity of ultra stable cw-lasers [14-18] to the microwave domain with

minute excess added noise [19-21], they also provide ultra-low phase noise microwave signals

[22-24]. Such signals may be of wide technological application in fields such as radar,

telecommunications, deep space navigation systems, timing distribution and synchronization

[25]. They have proven to be suitable as interrogation signal for atomic fountain clocks at the

quantum projection noise limit [26,27].

Among the different mode-locked laser technologies available, the Titanium Sapphire (Ti:S)

laser has traditionally been the workhorse of most optical metrology laboratories. Now

Erbium-doped fiber-based systems are currently evolving into very serious contenders,

especially when high reliability and very long term operation (several days or weeks of

operator-free continuous measurements) are crucial aspects. The main drawbacks of Erbium-

doped fiber-based systems, when compared to Ti:S lasers, have been the higher noise and

lower control bandwidth available. Laboratory systems with an intra-cavity Electro-Optic

Modulator (EOM) for improved control bandwidth have been demonstrated [28-30]. Highly

reliable systems, including an EOM actuator and exhibiting the high repetition rates suitable

for optical frequency metrology, are now available commercially with a built-in self-

referencing unit [31]. We present in this paper several characterizations of such a commercial-

core laser and demonstrate how we have used this information to understand and minimize

the residual noise of the system when phase locked to an ultra-stable cavity stabilized cw

laser, with a strong emphasis on low phase noise microwave signal generation. We start by

presenting the various parameters and actuators of the commercial-core optical frequency

comb. We proceed with a description of the methods used to measure actuators response

(transfer functions) and the relevant noise properties of our comb. We conclude by applying

these studies to the low phase noise microwave generation.

2. Optical frequency comb

A self referenced optical frequency comb is characterized, in the frequency domain by a set of

phase coherent optical frequencies νN=N×frep+f0, where N is the index of the mode, frep the

repetition rate (typically hundreds of MHz) and f0 the carrier-envelope offset frequency [1]. A

free-running optical frequency comb exhibits fluctuations of both frep and f0, as well as of its

average amplitude A. The state of the laser at a given time is therefore characterized by these

three distinct parameters. Note that, in principle, beyond these three “global parameters” one

could imagine that the phase and amplitude relation between the different spectral

components of the comb could also be changing over time. Due to the non-linear effects

responsible for the mode-locking mechanism, such variations have been shown to be minute

and, at best, have a very slow evolution [32]. We will therefore not take such effect into

account in this paper. A full characterization of the comb’s noise properties therefore requires

the measurement of three power spectral densities (PSD), of either phase or frequency noise

for f0 and frep, and of amplitude noise for A. Note that the effect of parameter A is usually

neglected in optical frequency metrology. In the context of low phase noise microwave signal

generation, where amplitude-to-phase coupling in the photodetection process is a strong

limitation [33,34], characterizing the amplitude fluctuations of the laser is however of utmost

Page 3

3

interest.

These three parameters (f0, frep and A), can be controlled via multiple actuators. In this paper,

we characterize a femtosecond comb system which possesses a piezo electric actuator (PZT,

whose purpose is to change the laser cavity length), allows fine tuning of the laser pumping

via adjustment of the current which drives the pump laser diodes, and has an EOM in the

cavity. The latter element operates mainly as a very fast voltage-controlled group delay

element in the femtosecond laser’s cavity. It therefore changes the repetition rate, although

with unavoidable coupling to the other two parameters (f0 and A). Ideally, the PZT modifies

only frep. However it is well known that a tiny misalignment of the laser cavity may influence

both f0 and A. Changing the pump power of the laser comb (via the current driving the pump

diode lasers) obviously impacts A, as well as both f0 and frep via complex mode-locked laser

dynamics [35-37]. A full characterization of the response of the comb to the various actuators

therefore requires the measurement of the nine transfer function matrix elements (three

actuators for three comb’s parameters) as shown by the following equation:

(

(

(

where Ipump is the pump diode current VPZT is the voltage driving the PZT and VEOM is the

voltage driving the EOM crystal. We have measured the nine transfer functions of this matrix

but we will focus on the subset which impact optical frequency metrology and optical-to-

microwave division process.

We proceed with a description of the methods we developed to measure these transfer

functions, as well as several noise properties of the femtosecond laser.

2. Transfer functions and noise measurements techniques.

Our commercial-core fiber-based frequency comb has a repetition rate near frep~250MHz,

which can be coarsely adjusted to within +/- 1MHz using a motorized translation stage on one

of the cavity mirrors. The available optical output power is about 130 mW. The laser has a

built-in f-2f interferometer unit which generates the offset frequency signal f0. A motorized

double wedge [37] allows coarse frequency adjustment of this quantity, which we maintain

near 70 MHz. To measure transfer functions, we use a two channel vector signal analyzer

(VSA - Agilent 89410A) which can operate up to 10 MHz Fourier frequency and has a with

built-in programmable source voltage output. The applied modulation is directed

simultaneously to the Channel 1 input of the VSA and the actuators of the laser which we

want to characterize (PZT, pump diode current or EOM) via a suitable actuator driver. Note

that these drivers need to be independently characterized, to insure they don’t impose

bandwidth limitation to the measurement. The quantity to characterize (frep, f0 or A) is

transformed into a voltage signal and fed to Channel 2 of the VSA. By programming the VSA

source output to generate a chirped sine wave, we measure the transfer function from Ch1 to

Ch2.

This allows obtaining the transfer function from the actuator to the comb parameter. The set-

up schematic is depicted in figure 1.

)

)

)

()()()

()()()

()()()

(

(

(

)

)

)

=

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

ω

j

V

V

I

HHH

HHH

HHH

f

f

A

EOM

PZT

pump

fVfVfI

fVfVfI

AVAVAI

rep

EOMPZTpump

repEOMrep PZTreppump

EOM PZTpump

000

,,,

,,,

,,,

0

(1)

Page 4

4

Fig 1: Setup schematic used to measure transfer functions of amplitude, f0 and frep vs laser pump current, intra

cavity EOM voltage and PZT voltage, as well as power spectral densities of the same quantities. f-to-2f: built-in

self-referencing unit, LNA: low noise amplifier, VSA: vector signal analyzer.

The key part of such an experiment is to design ways to transform frep, f0 and A into a voltage

signal with high sensitivity and minimal cross talk from the other two parameters of the comb.

Measuring A is quite straightforward. We inject a few milliwatts of light from the

femtosecond laser output (30 nm bandwidth around 1.55 µm) onto a fiber-pigtailed InGaAs

photodiode. The output of the photodiode is low pass filtered to remove the harmonics of the

repetition rate and, after low noise amplification, the output voltage is fed to the VSA. To

measure f0 fluctuations, we process this signal with a home-made fast tracking oscillator filter.

This device is composed of a voltage-control oscillator (VCO), which is phase locked to the

input signal with more than 3 MHz of bandwidth. At Fourier frequencies lower than this

bandwidth, the correction signal of the tracking oscillator (i.e. the voltage which controls the

VCO when the phase lock loop is operating) is proportional to the frequency fluctuations of f0

(the free running VCO noise is negligible compared to that of f0). To measure frep fluctuations,

it is convenient to increase the sensitivity by measuring N×frep, with N a large integer. By

beating the comb output with an ultra stable laser (linewidth<1Hz) [17] of optical frequency

νcw (near the 1.55 µm central wavelength of the comb), we have access to fb =N×frep +f0 - νcw.

This radio-frequency (RF) signal mixed with f0 leads to two sidebands, highly sensitive to frep

fluctuations (because of the large multiplicative factor N). We select, with a bandpass filter,

the sideband which is independent from f0 (since fb-f0 = N×frep-νcw). Locking a tracking

oscillator to this signal therefore leads to a voltage correction signal proportional to the

fluctuations of frep.

We use these techniques to transform A, f0 and frep into voltage signals to measure the nine

transfer functions and the noise power spectral densities of the optical frequency comb.

3. Results and applications

Once the comb is characterized, by the 3x3 transfer functions matrix, we can use this

information to improve the comb’s properties. Our work puts a strong focus on the context of

low phase noise microwave signal generation by photo-detection of the repetition rate’s

harmonics when the comb is phase locked to an ultra stable cw laser [23,24,27,38]. However,

the use of the transfer functions matrix goes well beyond what as appears in the following.

A first example of the use of the transfer function/noise characterization is illustrated in fig.2.

Page 5

5

10

0

10

1

10

2

10

3

10

4

10

5

10

-8

10

-7

10

-6

10

-5

10

-4

(2)

(1)

Frequency [Hz]

Laser pump diode current noise [A/Hz1/2]

early (noisy) pump laser current supply

low noise pump laser current supply

(a)

10

0

10

1

10

Frequency [Hz]

2

10

3

10

4

10

5

-160

-140

-120

-100

(3)

(2)

(4)

(1)

RIN [dBc/Hz]

Predicted for noisy pump laser current supply

Measured fo pump laser noisy current supply

Measured for low noise pump laser current supply

Predicted for low noise pump laser current supply

(b)

Fig 2: Plot (a): measured a spectral density of current noise for early noisy (1) (blue line) and low noise (2)

(black line) current supply. Plot (b): RIN of the femtosecond laser output with two different current supplies for

the pump lasers diodes. Curve (1) red (measured) and (2) black (predicted) are for the early (noisy) current

supply. Curve (3) green (measured) and curve (4) blue (predicted) are for the low noise current supply.

The first comb was equipped with a relatively noisy pump diode lasers current supply (Fig. 2

plot (a), blue curve (1)). By plugging the current noise of this current supply to the transfer

function we measured from the pump diodes’ current to the amplitude of the laser, we obtain

a predicted relative intensity noise (RIN) of the comb which matches very well the actual

measurement (Fig 2 plot (b), red curve (1) & black curve (2)). We can conclude from this that

the rather important noise of the early prototype current supply originates the excessive RIN

of the laser comb. Indeed, by using a low noise current supply as a replacement (Fig. 2 plot

(a), black curve (2)), we obtained a very substantial improvement of the RIN. Furthermore, by

comparing, for the low noise current supply, the predicted RIN (Fig. 2 plot (b), green curve

(3)) to the new RIN measurement (Fig. 2 plot (b), blue curve (4)), we can conclude that the

new current supply is not a limiting factor anymore.

10

3

10

4

10

5

10

6

10

-1

10

0

0

30

60

90

120

150

180

Phase [degree]

Transfer function HEOM,frep [Hz/V]

Frequency [Hz]

Fig 3: Transfer function

repEOMfV

H

,

from EOM voltage to repetition rate frep. Black is amplitude (left axis), red

plot is phase (right axis). The ultimate bandwidth available with such actuator is limited by the strong narrow

resonances near 900 kHz Fourier frequency.

By studying the RIN of the pump laser diode itself (when driven by the low-noise current

supply), we observed a RIN higher than expected from the current supply’s characteristic

alone. This excess noise explains the comb’s RIN for Fourier frequencies lower than 30-

50 kHz. Improving the performance may require to hand-select lower RIN pump laser diodes

and/or actively servo their output power. For Fourier frequencies higher than 30-50 kHz, the

RIN of the pump diode lasers does not explain RIN of the comb. Understanding this excess

Page 6

6

noise at high Fourier frequencies would necessitate further theoretical and experimental

studies.

A second example of transfer function application is to optimize the feedback loop when

phase locking the comb to an external reference (an ultra stable 1.542 µm laser in our

case).We give, in fig 3 and 6, the three transfer functions from the EOM control voltage to,

respectively frep, f0 and A.

From Fig 3, we deduce that the maximum available servo bandwidth when using the EOM as

an actuator to control frep is limited by sharp resonances observed around 900 kHz. As a

matter of facts, they produce rapid large phase shifts so large that they cannot be easily

compensated by control circuit design. We use proportional and multiple integration control

which acts on both the EOM and the PZT actuators (both act mainly on frep) to phase lock frep

to a cw ultra stable laser near 1.55 µm. Similarly to the technique we previously described for

measuring frep fluctuations, the beatnote between the cw laser and one nearby tooth of the

comb is mixed with f0 to provide a signal of frequency Nxfrep-νcw, independent of f0. By

mixing it with a reference from a fixed synthesizer, we produce an error signal which is sent

to a proportional-integral (PI) controller. This PI controller directly steers the EOM. The

EOM control voltage is further integrated and fed to the PZT. In this way frep is phase-locked

to the optical frequency νcw, with the EOM acting for Fourier frequencies larger than 6-

10 kHz, while the PZT controls the laser for low Fourier frequencies. Fig. 4 (a) shows the in-

loop error spectral density and Fig. 4 (b) the correction voltage spectral density applied to the

EOM when the phase-lock loop is running.

10

4

10

5

10

6

0

2x10

-4

4x10-4

0

1x10

-1

2x10

-1

Frequency [Hz]

In loop error [rad/Hz

1/2]

Integrated inloop error [rad]

(a)

10

3

10

4

10

5

10

6

0

1x10

-3

2x10

-3

3x10

-3

4x10

-3

0,0

0,5

1,0

1,5

2,0Integrated EOM voltage [Vrms]

1.7Vrms

Frequency [Hz]

EOM corr. voltage [Vrms/Hz

1/2]

(b)

Fig 4: Plot (a): Inloop residual phase error signal when phase locking the comb to a continuous ultra stable laser

near 1.55 µm. In red (right axis) the integrated phase error signal. Plot (b): rms correction voltage applied to the

EOM when the phase lock loop is running. In red (right axis), the integrated rms voltage applied to the EOM.

The decrease of the correction voltage below 6 kHz is due to the use of the PZT controlling the laser for low

Fourier frequencies (in lieu of the EOM).

The inloop error exhibits a resonance around 900 kHz and a servo bump near 400 kHz,

Page 7

7

consistent with the transfer function presented in fig. 3. By integrating the in-loop error up to

1 MHz, the total phase error is about 0.21 rad rms, resulting in 96.7% energy in the carrier. As

the total phase error is much lower than 1 rad rms, the comb is in the so called “ultra-stable

regime”, where each tooth of the comb has nearly the same spectral purity as the ultra-stable

cw laser used as a reference for the lock (once f0 is removed).

-40-30 -20 -10010 20 3040

-50

-40

-30

-20

-10

0

Beat-note signal power spectrum [dB]

ν-ν0 [Hz]

Fig 5: Radio frequency spectrum of the beat- note signal between the stabilized comb and an ultra-stable laser

near 1062.5 nm (Span 80 Hz, RBW 1Hz, 10 averages)

This was confirmed independently by beating the comb output near 1062.5 nm (obtained from

the same highly non-linear fiber used in the built-in f-2f unit) with a second ultra stable laser

(see Fig. 5)[39,40]. The integrated correction voltage is 1.7 Vrms (~6 Vpp maximum, verified

with an oscilloscope). When comparing to the 25 V output range of our fast control

electronics, this leaves room to optimize the EOM dimensions, using shorter crystals with

lower dispersion and, potentially, higher resonances which may allow larger control

bandwidths. A third example of the use of the transfer function is directly related to the

coupling transfer functions that are presented in Fig. 6. This figure represents the unavoidable

cross talks to, respectively, f0 and A, when we act on the EOM to control frep (and, ideally, frep

only). Understanding the dynamics of these two cross-talk responses (enhanced responses at

high Fourier frequencies in particular) is not straight-forward. We believe them to be caused

by minute misalignments of the EOM crystal, which couples, via complex laser dynamics, to

the polarization and amplitude of the pulse in the femtosecond laser’s cavity. Verifying this

hypothesis would require a thorough theoretical and experimental analysis which is well

beyond the scope if this paper. The measured transfer functions of Fig. 6 provide some useful

information about the limit of the EOM (in its current implementation) as an actuator to phase

lock frep.

Page 8

8

10

3

10

Frequency [Hz]

4

10

5

10

6

10

-9

10

-7

10

-5

(b)

Transfer function HVEOM,A [W/V]

10

3

10

4

10

5

Transfer function HVEOM, f0 [Hz/V]

(a)

Fig 6: Plot (a) transfer function

0

, fVEOM

H

(EOM voltage to f0 ). Plot (b)

amplitude)

0

,AVEOM

H

(EOM voltage to comb’s

By combining the correction voltage spectral density applied to the EOM when the servo loop

operating (Fig. 4(b)) to the transfer functions of Fig 6, we can deduce the EOM control-

induced fluctuations of f0 and A. The fluctuations of f0 are not important in the context of our

scheme, where f0 is uncontrolled and simply removed from fb by frequency mixing. Other f0-

removal techniques, including the one presented in [41], are likewise inherently immune to

such cross talk. Note that in experiments which independently phase lock frep and f0 as in Refs

[8,42,43]) may however need to account for those effects. On the other hand, the cross-talk

induced fluctuations of A are of pre-eminent importance in the context noise microwave

signal extraction by photo-detection of the femtosecond pulses. Indeed, amplitude-to-phase

conversion which occurs in the photo-detection process limits the achievable performance of

low phase noise microwave generation experiments [33, 34]. From the predicted cross-talk-

induced fluctuations of A, and a given amplitude-to-phase conversion factor related to the

photodetection process, one can deduce the achievable level of residual microwave phase

noise. For instance, in our case, for a microwave signal at 12 GHz generated from a 250 MHz

repetition rate laser, we assume a dφ/(dP/P)~0.03 rad amplitude-to-phase conversion factor of

the photodetection process. Here, φ is the phase of the microwave compared to that of the

pulse train and P is the average optical power on the photodetector. Note that substantially

higher conversion factors are normally without the fine adjustment of the experimental

parameters [33,34].

Page 9

9

10

1

10

2

10

3

10

4

10

5

10

6

-190

-180

-170

-160

-150

-140

-130

-120

(d)

(b)

(a)

Frequency [Hz]

Sφ(f ) [dB rad2/Hz] @12 GHz

(c)

Fig. Residual phase noise for optical-to-microwave frequency division operation of the frequency comb at

12 GHz. Curve (a) (red line): detection noise floor limit (shot noise at high Fourier frequencies) from ref [44],

plot (b) (black line): phase noise induced by EOM inloop error (i.e. due to limited bandwidth of the EOM

actuator).Plot (c) (green line) : RIN-induced phase noise. Plot (d) (blue line): EOM voltage to amplitude induced

phase noise. To obtain curves (c) and (d), we assumed typical 0.03 rad per relative amplitude change of the

amplitude-to-phase conversion factor in the photodetection process.

Fig. 7 presents the predicted residual phase noise limits obtained from RIN-induced phase

noise (in green) and EOM cross-talk-induced phase noise (in blue) value of the amplitude-

phase conversion factor. When comparing these predictions to the best noise floor limit we

measured for such a system with a repetition rate multiplication technique detailed in [44], it

appears that the cross-talk is not a limiting factor at the present level of noise floor limit. Note

however that if the amplitude-phase conversion factor were equal to 1 rad (typical worst case

scenario observed from [38]), it would constitute a limitation for Fourier frequencies in the

10 kHz-1 MHz range substantially higher than the measured noise floor limit. Furthermore,

the RIN-induced residual phase noise is also lower than the noise floor limit. Similarly, an

amplitude-phase conversion factor higher than 0.1 rad would induce a substantial limitation to

the performances of the system. On the other hand, it also appears from Fig. 7 that the main

limitation in the 100 kHz-1 MHz Fourier frequencies range is due to the residual inloop error

of the phase lock loop (black line). This limitation is ultimately linked to the still limited

actuation bandwidth achievable with the EOM. Further improvement of the EOM actuator

would therefore be necessary to achieve residual phase noise limits in the -160 dB(rad2/Hz).

Note that recently a 1.2 MHz bandwidth was achieved with a waveguide EOM [45] and we

could therefore expect a similar result for the future prototypes.

4. Conclusions

We have presented in detail the techniques used to measure the various transfer functions

which characterize an optical frequency comb equipped with various actuators. We have

shown some examples of how to use such characterization to deduce the performance limits

of the comb and improve its performance. When phase locking the repetition rate of the comb

to an ultra stable cw laser optical frequency, we reached the “ultra-stable regime”, where the

linewidth of each comb tooth is limited by that of the cw laser reference (and not the phase-

lock loop performances). In the context of low phase noise microwave generation by

Page 10

10

photodetecting the pulse train, we have made a prediction of the achievable minimum phase

noise from different possible noise sources. These prediction make us confident that we can

achieve residual phase noise in the -160 dB(rad2/Hz) level, close to the shot noise limit, with a

slightly improved EOM actuator. In addition we have shown how to use the dynamic

characterization of a frequency comb (via its actuators transfer functions) to understand and

improve an optical frequency comb’s performance.

5. Acknowledgments

We want to thanks J. Pinto for the valuable work for the electronic sub systems, and John Mc

Ferran for careful reading of the manuscript.

References

[1] Th. Udem, R. Holzwarth & T. W. Hänsch, "Optical frequency metrology", Nature, 416, 233-237,

(2002).

[2] N. R. Newbury “Searching for applications with a fine-tooth comb”, Nature Photonics, 5, 186–

188 (2011) and references therein.

[3] J. Reichert, M. Niering, R. Holzwarth, M. Weitz, Th.Udem, and T. W. Hänsch, “Phase Coherent

Vacuum-Ultraviolet to Radio Frequency Comparison with a Mode-Locked Laser”, Phys. Rev.

Lett. 84, 3232, (2000).

[4] S. A. Diddams, D. J. Jones, J. Ye, S. T. Cundiff, J. L. Hall, J. K. Ranka, R. S. Windeler, R.

Holzwarth, T. Udem and T. W. Hansch, “Direct link between microwave and optical frequencies

with a 300 THz femtosecond laser comb”, Phys. Rev. Lett. 84, (2), 5102–5105 (2000).

[5] R. Holzwarth, T. Udem, T. W. Hansch, J. C. Knight, W. J. Wadsworth, P. St. J. Russel, “Optical

frequency synthesizer for precision spectroscopy”, Phys. Rev. Lett. 85, 2264–2267 (2000).

[6] S. A. Diddams, L. Hollberg, L. S. Ma and L. Robertsson, “Femtosecondlaser-based optical

clockwork with instability < 6×10-16 in 1s”, Opt.Lett. 27, (1) pp 58–60 (2002).

[7] T. M. Ramond, S. A. Diddams, L. Hollberg and A. Bartels, “Phase coherent link from optical to

microwave frequencies by means of the broadband continuum from a 1-GHz Ti:sapphire

femtosecond oscillator”, Opt. Lett. 27, 1842–1844 (2002).

[8] Th. Udem, S. A. Diddams, K. R. Vogel, C. W. Oates, E. A. Curtis, W. D. Lee, W. M. Itano, R. E.

Drullinger, J. C. Bergquist and L. Hollberg, “Absolute frequency measurements of the Hg+ and

Ca optical clock transitions with a femtosecond laser”, Phys. Rev. Lett. 86, 4996–4999 (2001).

[9] X.Baillard, M. Fouch, R. Le Targat, P. G. Westergaard, A. Lecallier, F. Chapelet, M. Abgrall, G.

D. Rovera, P. Laurent, P. Rosenbusch, S. Bize, G. Santarelli, A. Clairon, P. Lemonde, G. Grosche,

B. Lipphardt and H. Schnatz. “An optical lattice clock with spin-polarized 87Sr atoms”, Eur. Phys.

J. D, 48, 11-17 (2008).

[10] N.D. Lemke, A. D. Ludlow, Z. W. Barber, T. M. Fortier, S. A. Diddams, Y. Jiang, S. R. Jefferts,

T. P. Heavner, T. E. Parker, and C. W. Oates “Spin-1/2 Optical Lattice Clock”, Phys. Rev. Lett.

103, 063001 (2009).

[11] M. Chwalla, J. Benhelm, K. Kim, G. Kirchmair, T. Monz, M. riebe, P. Schindler, A. S. Villar, W.

Hänsel, C. F. Roos, R. Blatt, M. Abgrall, G. Santarelli, G. D. Rovera, and Ph. Laurent, “absolute

frequency measurement of the 40Ca+ 4s2S1/2-3d2D5/2 clock transition,” Phys. Rev. Lett., 102,

023002, (2009).

[12] T. Rosenband, D. B. Hume, P. O. Schmidt, C. W. Chou, A. Brusch, L. Lorini, W. H. Oaskay, R.

E. Drullinger, T. M. Fortier, J. E. Stalnaka, S. A. Diddams, W. C. Swann, N. R. Newbury, W. M.

Itano, J. C. Bergquist, “Frequency ratio of Al+ and Hg+ single-ion optical clocks: Metrology at the

17th decimal place”, Science 319, 1808–1812 (2008).

[13] A. D. Ludlow, T. Zelevinsky, G. K. Campbell, S. Blatt, M. M. Boyd, M. H. G. de Miranda, M. J.

Martin, J. W. Thomsen, S. M. Foreman, J. Ye, T. M. Fortier, J. E. Stalnaker, S. A. Diddams, Y.

Le Coq, Z. W. Barber, N. Poli, N. D. Lemke, K. Beck and C. W. Oates, “Sr lattice clock 1×10-16

fractional uncertainty by remote optical evaluation with a Ca Clock”, Science 319, 1805-1808

(2008).

Page 11

11

[14] Y. Y. Jiang, A. D. Ludlow, N. D. Lemke, R. W. Fox, J. A. Sherman, L.-S. Ma, C. W. Oates

“Making optical atomic clocks more stable with 10−16-level laser stabilization”, Nature Photonics

5, 158–161 (2011).

[15] A. D. Ludlow, X. Huang, M. Notcutt, T. Zanon-Willette, S. M. Foreman, M. M. Boyd, S. Blatt

and J. Ye, “Compact thermal-noise-limited optical cavity for diode laser stabilization at 1×10-15”,

Opt. Lett., 32, 641 (2007).

[16] S. A. Webster, M. Oxborrow, S. Pugla, J. Millo and P. Gill, “Thermal noise limited optical

cavity”, Phys. Rev. A, 77, 033847 (2008).

[17] H. Jiang, F. Kéfélian, S. Crane, O. Lopez, M. Lours, J. Millo, D. Holleville, P. Lemonde, Ch.

Chardonnet, A. Amy-Klein, and G. Santarelli, "Long-distance frequency transfer over an urban

fiber link using optical phase stabilization," J. Opt. Soc. Am. B 25, 2029-2035 (2008).

[18] D. R. Leibrandt, M. J. Thorpe, M. Notcutt, R. Drullinger, T. Rosenband, and J. C. Bergquist

“Spherical reference cavities for ultra-stable lasers in non-laboratory environments” Opt. Express

19, 3471-3482 (2011).

[19] A. Bartels, S. A. Diddams, C. W. Oates, G. Wilpers, J. C. Bergquist, W. H. Oskay and

L.Hollberg,“Femtosecond-laser-based synthesis of ultrastable microwave signals from optical

frequency references”, Opt. Lett., 30, 667-669, (2005).

[20] J. J. McFerran, E. N. Ivanov, A. Bartels, G. Wilpers, C. W. Oates, S. A. Diddams and L.

Hollberg,“Low-noise synthesis of microwave signals from an optical source”, Elect. Lett., 41,

650-651 (2005).

[21] E. N. Ivanov, J. J. McFerran, S. A. Diddams and L. Hollberg “Noise properties of microwave

signals synthesized with femtosecond lasers”, IEEE Trans. on Ultrason. Ferr. Freq. Contr., 54,

736-745 (2007).

[22] T. M. Fortier, M. S. Kirchner, F. Quinlan, J. Taylor, J. C. Bergquist, T. Rosenband, N. Lemke, A.

Ludlow, Y. Jiang, C. W. Oates and S. A. Diddams, “Photonic generation of low-phase-noise

microwave signals” arXiv:1101.3616v1 (2011).

[23] J. Millo, R. Boudot, M. Lours, P. Y. Bourgeois, A. N. Luiten, Y. Le Coq, Y. Kersalé and G.

Santarelli “Ultra-low noise microwave extraction from fiber-based optical frequency comb”, Opt.

Lett., 34, 3707–3709 (2009).

[24] W. Zhang, Z. Xu, M. Lours, R. Boudot, Y. Kersalé, G. Santarelli and Y. Le Coq, “Sub-100

attoseconds stability optics-to-microwave synchronization”, Appl. Phys. Lett. 96, 211105 (2010).

[25]J. Kim and F. X. Kärtner, “Microwave signal extraction from femtosecond mode-locked lasers

with attosecond relative timing drift” Opt. Lett. 35, 2022-2024 (2010) and references therein.

[26] B. Lipphardt, G. Grosche, U. Sterr, C. Tamm, S. Weyers and H. Schnatz “The stability of an

optical clock laser transferred to the interrogation oscillator for a Cs fountain”, IEEE Trans. Instr.

Meas. 58, 1258–1262 (2009).

[27] J. Millo, M. Abgrall, M. Lours, E. M. L. English, H. Jiang, J. Guéna, A. Clairon, M. E. Tobar, S.

Bize, Y. Le Coq and G. Santarelli “Ultralow noise microwave generation with fiber-based optical

frequency comb and application to atomic fountain clock”, Appl. Phys. Lett. 94, 141105 (2009).

[28] D. D. Hudson, K. W. Holman, R. J. Jones, S. T. Cundiff, J. Ye, and D. J. Jones, "Mode-locked

fiber laser frequency-controlled with an intracavity electro-optic modulator," Opt. Lett. 30, 2948-

2950 (2005).

[29] Y. Nakajima, H. Inaba, K. Hosaka, K. Minoshima, A. Onae, M. Yasuda, T. Kohno, S. Kawato, T.

Kobayashi, T. Katsuyama, and F. Hong, "A multi-branch, fiber-based frequency comb with

millihertz-level relative linewidths using an intra-cavity electro-optic modulator," Opt. Express 18,

1667-1676 (2010).

[30] F. Quinlan T. M. Fortier, S. Kirchner, J. A. Taylor, M. J. Thorpe, N. Lemke, A. D. Ludlow, Y.

Jiang, C. W. Oates, and S. A. Diddams “Ultralow phase noise microwave generation with an

Er:fiber-based optical frequency divider”, Opt. Lett. 36, 3260-3262 (2011).

[31] www.menlosystem.com

[32] N. R. Newbury and W C. Swann, "Low-noise fiber-laser frequency combs" J. Opt. Soc. Am. B 24,

1756-1770 (2007).

[33] J.A. Taylor, S. Datta, A. Hati, C. Nelson F. Quinlan, A. Joshi, S.A. Diddams; “Characterization of

Power-to-Phase Conversion in High-Speed P-I-N Photodiodes,” IEEE Photonics Journal, 3, 140-

151, (2011).

Page 12

12

[34] W. Zhang, T Li, M. Lours, S. Seidelin, G. Santarelli, and Y. Le Coq, “Amplitude to phase

conversion of InGaAs pin photo-diodes for femtosecond lasers microwave signal generation”, in

press Applied Physics B and arXiv:1104.4495v1(2011).

[35] H.R. Telle, B. Lipphardt, J. Stenger, “Kerr-lens mode-locked lasers as transfer oscillators for

optical frequency measurements”, Applied Physics B: Lasers and Optics, 74, 1-6, (2002).

[36] J. J. McFerran, W. C. Swann, B. R. Washburn, and N. R. Newbury, “Suppression of pump-

induced frequency noise in fiber-laser frequency combs leading to sub-radian fceo phase

excursions”, Applied Physics B, 86, 219-227 (2007).

[37] R. Ell, J. R. Birge M. Araghchini and F. X. Kärtner, “Carrier-envelope phase control by a

composite plate,” Optics Express 14, 5829-36 (2006).

[38] W. Zhang, Z. Xu, M. Lours, R. Boudot, Y. Kersalé, A. N. Luiten, Y. Le Coq, G. Santarelli,

“Advanced noise reduction techniques for ultra-low phase noise optical-to-microwave division

with femtosecond fiber combs”, IEEE Trans. on Ultrason. Ferr. Freq. Contr., 58, 886-889 (2011).

[39] J. Millo, D. V. Magalhaes, C. Mandache, Y. Le Coq, E. M. English, P. G. Westergaard, J.

Lodewyck, S. Bize, P. Lemonde and G. Santarelli, “Ultrastable lasers based on vibration

insensitive cavities”, Phys. Rev. A 79, 053829 (2009).

[40] S. T. Dawkins, R. Chicireanu, M. Petersen, J. Millo, D. Magalhaes, C. Mandache, Y. Le Coq, S.

Bize, "An ultra-stable referenced interrogation system in the deep ultraviolet for a mercury optical

lattice clock", Applied Physics B: Lasers and Optics, 99, 41-46 (2010).

[41] S. Koke, C. Grebing, H Frei, A. Anderson, A. Assion, G. Steinmeyer “Direct frequency comb

synthesis with arbitrary offset and shot-noise-limited phase noise”, Nature Photonics 4, 462 - 465

(2010).

[42] J. Jones , S. A. Diddams , J. K. Ranka , A. Stentz , R. S. Windeler , J. L. Hall , S. T. Cundiff

“Carrier-envelope phase control of femtosecond mode-locked lasers and direct optical frequency

synthesis ,” Science 288, 635 (2000).

[43] Th. Udem, R. Holzwarth, and T.W. Hansch “Optical Frequency Metrology” Nature 416, 233

(2002).

[44] A. Haboucha, W. Zhang, T. Li, M Lours, A. N. Luiten, Y. Le Coq, G. Santarelli, “An Optical

Fibre Pulse Rate Multiplier for Ultra-low Phase-noise Signal Generation,” Optics Lett., in press,

arXiv: 1106.5195 (2011).

[45] E. Baumann, F. R. Giorgetta, J. W. Nicholson, W.C. Swann, I. Coddington, and N. R. Newbury,

"High-performance, vibration-immune, fiber-laser frequency comb," Opt. Lett. 34, 638-640

(2009).