Nasa's Phoenix spacecraft has gone silent and is unlikely to communicate with Earth again. The probe has succumbed to the diminishing light and increasing cold at its Arctic landing site.
The robotic lab had been investigating the region's climate and geology, trying to determine whether the planet has ever been capable of supporting life.
Dr Tom Pike, from Imperial College London, has been working on the mission. He kept a diary for the BBC News website of his experiences.
WEDNESDAY 12 NOVEMBER - FAREWELL PHOENIX
Right up to the end, we were still trying to squeeze more data out of Phoenix.
Our last set of commands for the Phoenix Lander was to break off a beam of our atomic force microscope.
The atomic force microscope station is now sitting silently on Mars
Like a used match in a booklet of matches, we wanted to get the tiny silicon beam out of the way, so we could carry on getting images with a new clean tip.
This is a dangerous manoeuvre.
We deliberately drive a miniature knife into our delicate array of silicon tips, cleaving off one while leaving the others untouched.
The knife needs to be aligned to within a hair's breadth - beam breaking had us sweating even when we practised on Earth.
But we'd already done this once on Mars, and had even been able to take a picture of our first broken tip in its final resting place, on one of the microscope substrates.
Now, after taking the highest resolution images of Martian dust, the second tip in turn had come to the end of its life.
But before the commands to remove this tip could be completed, Phoenix has come to a sputtering halt, running out of energy under the fading sun of a Martian autumn.
If humans ever venture to these bleak Martian arctic plains again... I'd like to think they will bring Phoenix some spare batteries
Phoenix will likely remain frozen for evermore, its knife poised above the second silicon tip - now given an indefinite stay of execution.
It's only now sinking in that the first Martian microscope station, a shoe-box sized instrument sitting on the deck of the Phoenix, has completed its life's work. It was first conceived in crude sketches in my lab book eleven years ago.
During the following three years, the microscope emerged from labs in California, Switzerland, Arizona and Germany, only to end up literally sitting on the shelf at the Jet Propulsion Laboratory in Pasadena for three years.
The array of tips used to image with the atomic force microscope
Reborn for Phoenix, it gained contributions from Denmark and the UK, our tiny micro-machined substrates from Imperial College London designed to hold the Martian dust and soil.
Nearly three years ago, the microscope station was bolted to the deck of Phoenix, before Phoenix in turn was hoisted onto a Delta II rocket - its last hours on Earth before the ten month journey to Mars in the vacuum of deep space.
But that was nothing compared to the four minutes of terror as Phoenix entered the Martian atmosphere just over five months ago, as first its heat shield, then parachutes, and finally retro rockets brought it to a halt.
The microscope station already had its first set of substrates poised to capture its first sample - the debris pushed up in to the thin atmosphere by the retro rockets.
The first tip broken off on Mars
During the next five months, we've loaded up all our substrates with dust and soil, taking over a thousand optical microscope pictures and nearly one hundred atomic force microscope images.
We've been so busy operating the microscope station to get more images, we've hardly had a chance to study all we've gathered so far.
But it looks like we've captured a billion-year-old history of Mars in our microscopic images. We see what might be ancient glassy grains that could have disintegrated into the familiar red dust that cloaks the planet.
Only the atomic force microscope was able to see the finest of the dust. We're getting hints in these particles of signatures of past water, spied by instruments circling Mars hundreds of kilometres above Phoenix.
The microscope station is now silent; optical microscope sightless; sample wheel rotating no further; atomic force microscope frozen; all waiting for commands that will probably now never come.
Delivery of one of the samples to the microscope station
But this instrument has given us by far the closest look we have ever had of another planet.
I'd like to think that if one day humans ever venture to these bleak Martian arctic plains they will bring Phoenix some spare batteries.
And if the microscope station fires up once more, I hope they can send those last long-delayed commands - to break off the beam and get some final images to add to the collection.
FRIDAY 15 AUGUST - A SPECK OF DUST
When we first conceived of a microscope for Mars nearly twelve years ago, it was to help search for hazards for future human exploration of the planet.
Combined with an analysis from a wet chemistry laboratory, we hoped to alert Nasa to possible environmental threats to astronauts from the Martian dust and soil.
The AFM scan of the Antechamber revealed a single dust particle
A mission to Mars requires astronauts to remain months on the surface before a return launch to Earth.
During that time it is inevitable that some dust and soil will get inside their living quarters. On Earth, exposure to small amounts of very fine dust can be extremely damaging - it is the tiny diameter of asbestos fibres that allows them to get deep into the lungs.
Being able to image such small particles drove us to combine an optical microscope with an atomic force microscope (AFM). We called our set of instruments MECA, for Mars Environmental Compatibility Assessment. It was built for the cancelled Mars 2001 Lander.
MECA is now operating on Phoenix, and its role has changed. Now its aim has broadened to study Mars as a planet.
The name has changed as well: still MECA, but now this stands for the Microscope, Electrochemistry and Conductivity Analyser. But there are still plenty of reasons to study the dust of Mars apart from the effects on humans.
The dust is everywhere. As Mars has almost certainly not seen rain for billions of years, the dust has built up as a fine coating on everything and soil may be hundreds of metres thick.
It stays lofted in the sky giving the distinctive orange-pink backdrop to photographs taken on the planet, and giving sunsets on Mars a ghostly blue glow. Seasonal dust storms can envelope the whole planet, drastically reducing the Sun's heating of the surface.
It is impossible to properly understand Mars without studying the dust, but we are the first to try to take an image of individual dust particles.
The silicon substrate is patterned to capture the Martian dust
We know these particles must be small, at best a few micrometres, millionths of a metre, across. A couple of weeks ago we confirmed that our AFM has sufficient resolution to see the dust. Now we have to hold a single speck still while the AFM scans it.
The silicon substrate etched with tiny pits over two years ago was designed to do this job. This 3mm-diameter disc of silicon was carefully processed with these "nanobuckets" by Sanjay and Hanna in Imperial College's nanotechnology clean room back in London. We are going to see if our nanobuckets will work.
The robot arm has sprinkled a perfect sample onto our nanobucket substrate. After taking an image with the optical microscope we have to select a target area we'll scan with the AFM.
We're not looking for the actual particles: they're too small to be seen in this image. Instead we're looking for where there are no larger particles that could get in the way of the AFM while it scans.
All our targets for study on Mars should have a name. We've chosen fairy tales as a rich source of labels for our targets. Our nanobucket substrate is number 67, so I decide to rename it "The Twelve Hunters," the title of the Brothers Grimm's 67th tale.
This is not one of the Grimm's better-known stories so I have to read it to find a target name for the AFM scan. In some ways it's a typically implausible story of thwarted love between a prince and princess.
But as I read on I feel a chill. The prince tries to unmask the princess, disguised as one of twelve hunters, by demanding she leave her bedroom across an unlit antechamber whose floor is strewn with dried peas.
The Phoenix team is based at an operations centre in Arizona
If she doesn't have the firm step of a hunter, the peas will be scattered across the antechamber and she will be unmasked. But the princess hears of the prince's test, treads firmly leaving the peas undisturbed and eventually marries the prince in a classic fairy-tale ending.
So our target area of nanobuckets on the Twelve Hunters is named "Antechamber". We send the coordinates up to Phoenix to command the sample wheel to rotate Antechamber in front of the AFM and take a scan.
The next day we're back in the downlink room, waiting for the AFM scans to return. It's taken more two-thirds of the primary mission of Phoenix to get to this point. First we had to get the sample wheel and optical microscope to bring back images of the soil.
Then the AFM was brought into operation to produce good scans of the test substrates. This will be our first attempt to image any Martian dust particles in the nanobuckets.
As the data comes down and we decode it, we first see that the AFM has imaged the nanobuckets of Antechamber.
There are several visible as dark circles in the scan. As I plot the surface in three dimensions, I see that one of the nanobuckets right at the edge of Antechamber looks different. There, lodged to one side, is what looks like a particle.
The AFM scans both forwards and backwards, producing two images as it scans. I immediately check the reverse scan and see the same feature in the image.
Our particle has held still in the nanobuckets of Antechamber as sure as any of Grimms' peas. Urs, Sanjay, Mike and I look at each other. Is this the fairy-tale ending?
As a documentary TV camera crew peer over our shoulders, we measure the size of the particle. At one micrometre across it's just about as small as we expected. We are the first people to see the individual specks of dust on Mars.
We're not finished. Next we have to work on building up a portrait gallery of the Martian dust particles. We also want to look at what collects on our substrates if we leave them outside the MECA enclosure overnight.
But before we continue our work, we invite the entire Phoenix team over to the Microscopy House to celebrate. We've lived this Martian tale for seventy sols together; many more if you include all the training sessions in Tucson. We want to thank the whole team for helping give this chapter a happy ending.
WEDNESDAY 6 AUGUST - NOTHING TO HIDE?
It was more than a month ago, on sol 30, that the wet chemistry laboratory (WCL) on Phoenix first detected what looked like perchlorate in the Martian soil.
We've waited until sol 72 to report this finding, and even this has been a somewhat reluctant announcement. To many people, the question will be what did we have to hide?
The wet chemistry lab beaker contains sensors in its walls
The answer highlights a fundamental difficulty of doing good science under the gaze of the public.
When the science relates to the possibility of life on another planet, that gaze can become very intense. The data from WCL that streamed back to Earth was not immediately ready for such public scrutiny.
First, the results from WCL needed careful analysis - the sensors used to identify the chemistry of the Martian soil don't directly give us a list of ingredients.
We need to look at how any one chemical might affect more than one sensor - it's more like a Sudoku puzzle with chemicals rather than numbers, and, similarly, not a great spectator sport.
As soon as the data arrived from Mars, the WCL team were busy in the laboratory here in Tucson, using a twin of the instrument on Mars to understand how all of the readings from the sensors could be explained.
Some results quickly emerged - the soil was mildly alkali. This was announced at a press conference the very next day. Others, like the confirmation of perchlorate, took a little longer.
Secondly, WCL is not the only instrument on Phoenix that can look at the soil chemistry. TEGA, the thermal evolved gas analyser, bakes the soil and can identify the mass of the molecules that are given off.
However, TEGA can't look for all the possible chemicals, and it wasn't even looking for perchlorates when it cooked its first sample. We really wanted to wait until TEGA had a chance to confirm the perchlorates before going public.
The wet chemistry lab is looking at soil chemistry on Mars
Thirdly, it's always wise to get an outside opinion before presenting an important result. This is the basis of the peer-review process, a much used phrase in scientific circles which just means that an independent expert in the field should take a close look before you go public.
Completing all of these steps would take weeks or months. Skipping a step or two can lead to embarrassment at best. At worst, if faulty results are presented as correct, research can be knocked back years.
Against this measured pace of careful confirmation is the strong desire to show our results to the world. A space mission is a rare opportunity to share the excitement of scientific discovery with a worldwide audience.
We've had camera crews regularly filming in the SOC, and have given regular press conferences as well as numerous radio and TV interviews. The Phoenix science team is happy to present the edited highlights.
Transmitting a live commentary, complete with our speculations and initial and often faulty analyses, would be a very different proposition.
Martain soil is delivered to the WCL using the lander's scoop
The journalists covering Phoenix had been expecting more complete WCL results for many weeks. They quite reasonably guessed that the delay must mean we were working on something important.
It was reported the White House was briefed that WCL had returned results relating to the potential for life on Mars. By last weekend, the internet rumour mill, fuelled by online news sources and bloggers, was in overdrive.
Even the microscope station, which shares the enclosure and some electronics with WCL, became part of a cover up: according to one source we had imaged alien droppings!
Peter Smith, who leads the Phoenix Team, was in a tricky position. Keeping quiet about the perchlorate would just fuel speculation that was getting wilder by the day.
But the detailed work on the perchlorate analysis had not been completed and TEGA had been unable to make a positive identification, though with a different sample.
Peter made the difficult decision to make a preliminary announcement on what we had found to date.
Rather than a confident declaration of our findings, this announcement came with a science health warning - this was a work in progress, and as such, the conclusions could shift.
Usually, any scientist feels flattered when a journalist calls to talk about their work.
But for a brief period in the last week we had to give that most uncomfortable stock response: "No comment."
At least after this week's announcement, we can get on with the mission without looking like we've got something to hide.
FRIDAY 1 AUGUST - UNDER PRESSURE
The heat is on...
Since I returned to Tucson more than two weeks ago, we have been struggling to bring the atomic force microscope into action.
The AFM should be able to zoom in well beyond the resolution of the optical microscope so that we should be able to see the structure of the individual dust grains on Mars.
The AFM has eight beams, each with a pointed tip used to scan samples
That should give us clues to whether liquid water might have been responsible for weathering the rounded grains we see.
The AFM could be the most complicated instrument sent to operate on the surface of another planet.
At its core is an exquisitely fine tip which scans over the sample, carefully following the contours to produce a three-dimensional image of the surface. On Earth it is possible to even map out individual atoms on a surface.
The tip is at the end of a tiny beam of silicon shaped like a miniature diving board. This silicon diving board has a distinctive twang, way above the acoustic frequencies we can hear, but the AFM can use any slight shift in the tone of the twang to keep the tip scanning just above the surface of the sample.
These tips can wear down, so we've sent eight of them, each at the end of their own silicon beam.
This all works very well in the laboratory here on Earth, but on Mars we have a major complication: the temperature swings between -20C in the early afternoon to -80C at night.
As the temperature drops, the distinctive tone of the twang rises as the silicon stiffens. We're looking for tiny differences in tone while we scan, but we're being swamped by changes in temperature.
The team had all been working towards this image
It looks as though we might have sent a rather complicated thermometer, rather than a microscope, to Mars.
Of course we can't expect the polar regions of Mars to mirror the temperatures of a laboratory, and we've built a chamber so that we can run the AFM at the temperatures and pressures we might expect on Mars.
But short of putting the entire spacecraft in a massive chamber, it's never possible to duplicate all of the conditions of a mission. Very often it is difficult to be sure how well an instrument might perform before it reaches its destination.
We've worked out a scheme to turn on the electronics of the AFM early. This should heat up the inside of the black box of our microscopy station to what we hope is a stable temperature before we move our samples up to the AFM and start scanning.
We send up the revised code to the spacecraft for the next sol's AFM operations. As the downlinked AFM images arrive on Earth, it looks like this might be helping: the scans still need considerable improvement, but we're moving in the right direction.
While we're crowded round the computer in the downlink room, the figure of the mission manager looms in the doorway - our heating has pushed one of our temperature sensors on the spacecraft into the red, above 65C. It's just by a couple of degrees for less than a minute, but he wants an explanation.
More importantly he wants to know how we're going to avoid doing it again. We're in a bind: although it looks as though we might have found a way to run the AFM, we could be risking our instrument at the same time.
As I pore over the temperature profiles, it becomes clear that the problem is not with our heating of the electronics, it's the final burst of current as we use the motors to approach the AFM. We need to find a way to cut back on the motor power.
Tom, Sanjay and Urs celebrate their success
Until we do, the AFM is banned from working at the hottest part of the Martian day, the very time the temperature is most stable.
Rushing back to the laboratory, we use the copy of the flight AFM in an environmental chamber to work out a revised way of using the motors.
By running the motors more quickly, followed by regular pauses, we can drastically cut back on our power. While the code for this new approach is reviewed, I model the temperatures we might get - we'll keep well below the 65C red line. The AFM is put back in the schedule.
The next day I feel a rather unpleasant mixture of eagerness and anxiety as we wait for the AFM data to come back.
This time our images are expected to arrive on Earth later than most of the other instruments' data. I have time to think through our precarious position. We can't expect to be given too many more opportunities to get the AFM running properly. The heat really is on - this might be our last chance.
As the microscope team waits in the MECA downlink room, the TEGA team down the corridor explodes - their instrument has finally produced the confirmation we were looking for - Phoenix has identified water.
Their hard work has been rewarded, but I feel I can only take a sip of the champagne to celebrate their much deserved success.
With the SOC still buzzing, we see the AFM images trickle down. The first scan looks stable - as we hoped our temperatures must have kept flat while we imaged. I start plotting the data in three dimensions.
I can make out the details of the calibration grid we've imaged, an array of tiny square blocks in a chequerboard pattern. You could fit one million of these blocks on the head of a pin. The scan is more than stable, it is superb. In fact it is better than any scan of this grid we've made on Earth.
We stare at this image, allowing the slow realisation of what we have achieved sink in. However, this is only a grid from Earth, not a sample from Mars, and we need to start imaging particles.
We started late and we're running out of time - we're now more than two-thirds of the way through the original mission.
Perhaps spurred by TEGA's result, Nasa gives the announcement we were all hoping for: they're willing to extend the mission beyond the 90 sols to the end of September.
Imaging particles with the AFM was never going to be straightforward, but at least we now have more time to tackle the challenges ahead.
MONDAY 21 JULY - CHANGING TIMES
After Phoenix landed on Mars back in May the science team all felt a very sharp stick in our backs prodding us on - each day of surface operations might be our last.
Hanna has been making ice-soil mixtures
The microscope was the first instrument to get a sample delivery from Mars - it came raining down as we landed. We were able to send microscope images back to Earth within the first ten Martian days, or sols. Our impatience was rewarded..
Now we are 54 sols into the mission and, as I return to Tucson, a new mood has descended on the science team. There is an air of patient concentration as we work out ever more challenging ways to study this corner of the Red Planet's polar plains.
The science team has been working out the best way to get a sample of ice into the TEGA instrument, where its ovens can boil it and analyse the chemical composition.
Hanna has been making ice-soil mixtures to see how long we have to image the mixture before the ice vapourises. Urs and I have been working on how to use the atomic force microscope (AFM) to image the particles we've collected so far.
While I was away, Urs and the microscope team took the first steps and used the AFM to successfully image one of our calibration substrates. The AFM was able to pick out individual grooves one thousandth of a millimetre across, invisible to our optical microscope.
Hours of preparation
The particles, though, are much more difficult to image. They tend to be pushed about by the sharp tip of the AFM, frustrating our attempts. On Earth, we'd be able to turn a few knobs to minimise the forces between the tip and the particle.
For Phoenix, each "tweak" of the computer code takes hours to prepare and the results from Mars aren't known until the next sol. We also have to book ahead to be able to run our microscopes on Phoenix.
What takes just a few minutes to run on the testbed in our laboratory in Tucson takes several sols to complete on Mars. A good deal of patience is a requirement for any member of the science team.
Scientists are trying to get a sample of ice into one of the instruments
The sense of urgency has not disappeared altogether. Our mission will end as the Martian summer draws to a close.
Currently, though, we are "power positive". Phoenix often needs some time to cool down after a strenuous sol of activities. We even have enough power to keep Phoenix operating continuously for an entire sol.
This gives us an opportunity to see how the frosting of the soil and ice at the bottom of our trench changes under the weak sunlight of a Martian polar night.
The spacecraft crew in Denver have developed a deeper understanding of how our solar arrays are capturing power while the instruments burn it. They estimate that we should still be "power positive" until November, giving us at least a 150-sol mission.
Phoenix may continue to have plenty of power during the day shifts for a while to come, but the science team will soon stop working Martian nights.
This is the time that we're kept busy analysing the results and preparing the code for the next sol. For the second time in the mission, the nearly twenty-five-hour Martian day is re-synchronising our sleep cycles to Tucson time.
Back to normality
By the end of the week, we'll slide over to Earth time and stay there. It will make running the mission more complex: we'll have to plan the next sol's operations without the benefit of the most recent data returned from Mars.
But it will allow at least the local members of the team to have something approaching a normal life.
It has been quite an experience being one of maybe a hundred people on Earth whose body clocks have been locked to the rotation of another planet.
The Phoenix science team is a very international group and it's been easy to think of our collective home being Earth. But for the last two months we've felt part of an even larger entity, with our waking hours determined by the gyrations of the Solar System.
But before we desert Mars time, we have time to hold an early morning party before Hanna leaves us briefly. She's about to get married back in England, and for this wedding the planets are definitely in alignment: Hanna's last Martian shift leaves her synched to British Summer Time.
TUESDAY 8 JULY - FUNDING MARS
I'm back in London for ten days, catching up with the last duties of the term at Imperial College, reminding my family of my existence and attending meetings about the upcoming set of space missions.
ExoMars, unlike Phoenix, will be hunting directly for life
One of these meetings concerns ExoMars, the 2013 European launch to Mars and what we hope will be the first non-US mission to operate on the surface of the Red Planet.
ExoMars, unlike Phoenix, will be hunting directly for life and that is a part of the reason that it will be three times as expensive as Phoenix.
The instruments aboard ExoMars that might identify life are incredibly sensitive - any contamination from Earth could lead to a false positive. So a programme of sterilisation has to be built into the development of much of the other hardware on the mission.
Back in 1976, the two Nasa Viking landers had to go through the same process. The spacecraft were baked at 120C for seven days in giant ovens just before their launch to destroy any terrestrial life.
The microseismometer on ExoMars can pick up marsquakes
In 1974, for every $10 the US was spending on protecting its environment, $1 went on safeguarding the environment of Mars from microbial contamination from Earth.
Viking set the standard, and the cost, of the very high levels of biological cleanliness required for any future mission searching for Martian life. The price tag of ExoMars reflects this cost.
ExoMars is also an ambitious mission. For Viking, as well as Phoenix, only samples within reach of the scoop could be analysed.
The life detection instruments of ExoMars will be mounted on a rover able to roam many kilometres and drill down 2m to find samples. That represents a massive increase in the volume of Mars that can be searched, and a further substantial increase in the price tag.
Of course, there is a distinct possibility that ExoMars will not identify life, at least in the areas it can search. But the mission will not just be concentrating on this quest. It is planned to include on the mission a stationary set of instruments to look at the atmosphere and interior of Mars.
One of these is a microseismometer - a very sensitive vibration sensor we're working on at Imperial to detect marsquakes. This should help us see for the first time deep into the internal structure of another planet.
Again, all these instruments, with the power and data systems required to run them, cost money.
Both Viking landers were baked to sterilise them of microbes
Considering how many more capabilities are included on ExoMars, it is not so surprising that its current budget is three times that of Phoenix.
Neither is it surprising that there is considerable anxiety among the potential British contributors to ExoMars as news of a 25% cut in UK funding for the programme emerges.
Both Mars Polar Lander and Beagle 2 bear testament to underfunded projects that used up their budgets as well as large chunks of people's careers only to end up as scrap metal on Mars.
By contrast, Phoenix has already been a success in a number of ways - the landing was executed superbly, it is the first mission to touch the ice on Mars, and its instruments are giving us the closest look at the appearance and chemistry of the Martian soil.
It is also the cheapest mission to make it all the way to the surface of Mars. I'm sure we could have axed a quarter off the $460m cost of Phoenix, but I don't think I'd now be booking my flight back to Tucson to share in the discoveries of the remainder of its mission.
TUESDAY 1 JULY - CHEMISTRY OR LIFE?
Phoenix does not have the tools for a thorough investigation of current life on Mars.
We are investigating whether there is at least the potential there. By contrast, over 30 years ago the two Viking landers each carried a set of four experiments on board designed specifically to detect life on the Red Planet.
The Viking landers carried experiments to detect life
One of these experiments mixed a scoopful of the soil with nutrients and looked to see if any of the carbon in these nutrients was given off as gas.
The results were at first astounding - something in the Martian soil did indeed seem to digest the nutrients and expel gas.
But the other three experiments were negative. Was Viking seeing life, or chemical reactions between the nutrients and the soil itself?
On the deck of Phoenix, on one side of the black box that houses the microscope station, are four chemistry experiments waiting to run. This is the Wet Chemistry Laboratory (WCL), essentially four test tubes that will mix the Martian soil with pure water from Earth and see what we can extract.
We will be recreating in these test tubes what we believe would have regularly taken place billions of years ago when Mars was a much warmer and wetter planet. WCL might also give clues to why the soil might have appeared to harbour life in that one Viking experiment.
Inserted into each of the WCL test tubes are a number of sensors - one is an electronic litmus paper which will tell us how acidic or alkaline the soil is. Taken together, the results from all these sensors will give us the best information yet on Mars' surface chemistry.
Phoenix has delivered its first sample to the wet chemistry laboratory
In particular, if the soil is alkaline this may go a long way to explaining the Viking experiment. Fifteen years ago, one of my Phoenix colleagues, Richard Quinn, worked out that just such a soil would explain the Viking results.
He is now in the Phoenix science operations center (SOC), anxious to get soil into the first of these test tubes.
It has taken a while - nearly 30 days since Phoenix landed - to make sure we are ready to go with WCL.
It requires a complex sequence of releasing water from a tank, bringing soil into the test tube, stirring the mixture up to make sure we release all the soluble chemicals, and, finally, taking a slew of measurements from the resulting brew.
We have only one shot at it, and we want to make sure the spacecraft follows these instructions to the letter.
The first sample for WCL comes from the top few centimetres of the Martian soil. The optical microscope has already indicated this very top surface uniformly coats Mars. Near the Martian equator, Viking probably looked at a very similar scoop of material to what Phoenix has dug up from the "arctic" plains.
The data streams down, filing up our computer screens with scans from all the sensors. There's a quiet smile on Richard's face as he concentrates on just one of these lines - the soil is just about as alkaline as he expected.
It looks like one of the biggest mysteries from previous missions to Mars has been cracked by Phoenix. It was the chemical reactions of nutrients with the chemistry of the soil, not life, that Viking saw back in 1976.
FRIDAY 20 JUNE - HUNTING FOR ICE
It was the promise of ice that brought Phoenix to the arctic plains of Mars. The mission was funded on the expectation that we would find ice.
While the proposal for Phoenix was being written, the Mars Odyssey Orbiter was sweeping its instruments over the surface of Mars 300km (186 miles) below.
One of these, a gamma ray spectrometer, was looking for one very particular signal, the distinctive signature of hydrogen.
The only plausible source of hydrogen would be water in the form of ice, frozen below the surface of Mars.
The spectrometer did more than just find ice; as the orbiter circled Mars, it built up a complete map of how much ice there was, and how far below the surface we were likely to find it.
On the northern plains of Mars, the Vastitas Boralis, it would be just a few centimetres down. The proposal for Phoenix had a landing site and a goal: to dig down and touch this ice.
Phoenix landed in the blue ice-rich area (top left), near Mars' north pole
I'm supposedly having a day off today. I already have my eye on upcoming missions both to Mars and the Moon, and I need to do some work on preparing the next set of instruments to send into space.
Rather than microscopes, next time I hope it will be seismometers. Instead of zooming in on the individual dust grains, in 2013 I'll be looking for "Marsquakes" using a very sensitive vibration sensor we're developing at Imperial.
To probe deep into the Mars, Tom is also working on "Marsquake" sensors
This will tell us about the deep structure of Mars, and should be the first time we learn about the inner workings of a planet other than our own.
It's difficult to shake the feeling that I'm somehow being unfaithful to Phoenix. Even if Phoenix lasts beyond the 90 days it is expected to survive, it won't make through the bitter Martian polar winter.
Every day of this mission is precious, and there's always new data to look at. I'm itching to get back to the science operations center (SOC).
Early one morning I drove quickly to the SOC, eager to see what has come down in the latest transmission from Phoenix.
An overnight late pass of an orbiter overhead has given the lander a chance to send back some of the pent-up images it had been storing from a day before. I get to have a first look at the some images of the trench.
We've seen white material at the top of the trench; if this were ice, it is showing little change.
But if this was part of a much larger layer of ice we might not see much change because it would be kept cold by the rest by a much larger hidden mass below.
What I want to find are small white particles. If ice, these would turn to gas, or sublime, in the thin Martian atmosphere. In other words, the particles would shrink and disappear.
Small ice particles in the trench are disappearing into the atmosphere
I try to coax some detail out of the shadows.
I'm fighting against image compression because in order to get as much data as possible back to Earth, we shoehorn the images into the smallest possible space.
It's the dark shadows that lose most of their detail.
But in the deep shadows, I start to make out the outline of some small white shapes.
I compare these outlines to the image from the day before - there's been no digging in between yet there has been a change: the particles are shrinking.
At the end of every Mars working day, we have a chance to present what we've found to all the Phoenix mission scientists. I convince many, but not all.
Confirmation, though, should be simple. I work out that in another couple of days the particles should have disappeared altogether.
It's been just over six years since the Mars Odyssey Orbiter first spotted water. I can be patient for a few more days.
Yesterday, we got a chance to image the trench again. The white particles are gone. We've found what we came for.
SATURDAY 14 JUNE - MARTIAN LABOURS
It's been more than 30 years since a spacecraft has grabbed hold of the soil of Mars. In 1976, the two Viking landers sampled the surface for signs of life.
The scoop delivered soil to the lander's microscope
Since then there have been just three rovers on Mars, two of them still trundling around.
For them, the soil has been a road surface at best, a sand trap at worst. It has not been a substance to pick up and handle.
Phoenix has a much more sensual relationship with the planet's surface. It is digging furrows with its robot arm, stopping to collect a scoopful to bring back to the instruments waiting impatiently on its deck.
But we have little idea what the soil is like, and we've been fumbling while we try to understand what we're handling. There's very little to help us get a feel of the soil.
The three orbiters circling Mars have sent back pictures at a level of detail never seen before of the surface of the planet.
We've looked down on ice-filled craters, canyons that dwarf even the largest on Earth, and glimpsed Phoenix itself swinging down on its parachute.
That last image now forms the backdrop to nearly every laptop screen at the science operations center (SOC). But the orbiters can only give the most tenuous clues as to how the soil might really feel.
We suspect that, unlike the Earth, the make-up of the soil on Mars varies little - the planet is wreathed in the particles of innumerable global duststorms. That dust, though, has settled in very different environments.
At the poles it lands on ice caps, soon to be buried every Martian winter by ice fall. At the equator it becomes part of a dry desert.
White material at the bottom of the trench could be ice
Where Phoenix has landed, it seems to have a complicated relationship with the ice we think we've spotted just below the surface.
Our soil is unpredictable. First we dig up clumps that disintegrate overnight in the scoop. Then the robot arm tries to dump the first sample into the oven where it will be baked and analysed by a mass spectrometer, an instrument called TEGA.
But the generous serving sits obstinately on the sieve that should just prevent pebbles jamming up the works. Even shaking the sieve has no effect. The TEGA team is becoming exasperated by the soil.
We're next with a delivery to the microscope and decide on a more delicate approach - sprinkling the soil on to our substrates.
The scoop has a vibrator built into it, designed to free up the hard ice and frozen soil we might collect later. During testing back in March, we worked alongside the robot-arm team using the copy of the spacecraft in Tucson. We found out how to use the vibrator to dust our substrates with material from a tilted scoop.
We want just enough soil to coat our substrates, but not so much that we can't make out the individual particles. But in our test in Tucson we could only use our best guess for the soil on Mars.
Perserverance finally paid off for the Phoenix team
It has now taken three days of testing on Mars, sprinkling soil on the top of the black box housing the microscope, before we've convinced ourselves that we can deliver actual Martian soil.
We command the final sprinkle, followed by moving our sample wheel holding the substrates inside the box and round to the field of view of the microscope.
We hope this final movement, rotating the substrates from horizontal to vertical, will leave just the right amount of soil to peer at with the microscope.
At three the next morning we're back in the SOC, waiting again for our images to return from the spacecraft. We're much more confident that the microscope station will perform as commanded.
My doubt now is whether the Martian soil will cooperate. All eyes are on us - the mission will only proceed if we can be certain the microscope got the first sample of Martian soil.
Until we give the signal, the robot arm hangs frozen over the microscope enclosure.
The first images show the scoop and our substrates poking out of our box - with what looks like just the dusting we were looking for.
But it is not until Sanjay again decodes the incoming data that we see our first microscope images. The soil is scattered perfectly across the substrates. It's a Go.
Our excitement is mirrored by the TEGA team - on the seventh attempt at shaking their load, the obstinate soil has finally collapsed into their awaiting oven.
We congratulate each other like proud new parents - after a few anxious days of labour we've both got our deliveries from the surface of Mars.
SUNDAY 8 JUNE - THE DUST ON MARS
The first sample we want to look at with the microscope is an insurance policy. Even if the robot arm is for some reason unable to reach down and scratch out a sample for us, we know the lander's retro-rockets (fired at the Martian surface to slow the craft's descent on landing) would have done the trick.
Phoenix has returned its first microscope images
We landed in a cloud of dust as Phoenix lowered itself onto the surface of Mars. We just had to be able to collect some of it.
So a month before landing, while still millions of kilometres away from Mars, we moved our first set of test substrates out of the microscope station enclosure ready to catch the debris thrown up during landing.
These substrates are the discs we use to hold the dust and soil of Mars in front of our microscopes.
We then wrote the spacecraft computer sequence to rotate the substrates back into the enclosure, focus the microscope and take the red, green and blue images to give us the colour pictures of the dust we have managed to collect.
All is set, and there's little else to do now but wait for the sequence to run. We bundle the kids into the car and take advantage of a couple of days off to head up to the Grand Canyon.
The news of Phoenix precedes us as we stay overnight on the way in Flagstaff. As we eat a late supper we're congratulated by our fellow diners on a mission that Arizonans have obviously taken to their hearts.
But until we have received our first images of dust, I still feel like a tourist rather than a scientist on Mars.
It's a truism that no camera can capture the vast scale of the Grand Canyon. But as we amble along the canyon's North Rim, it is the timescale as much as the immensity of what we are gazing at that is staggering.
The Grand Canyon was created in about 20 million years
It took 20 million years for the Colorado River to carve out the scene in front of us.
But even that is an instant compared to the four billion years that Mars has had to grind its own surface down to a fine dust.
We believe Mars was a wetter and warmer place early in its history.
But it dried out long ago, and since then, the dust has been whipped into innumerable sandstorms - often enveloping the entire planet.
It is these tortured individual particles that we hope to see for the first time.
I'm back in the science operations center (SOC) in Tucson just in time for the data to arrive back on Earth. This time, the transmission stutters and we wait impatiently for the full pictures to come down.
No one has looked at the dust of Mars at this resolution before, and now I gasp as the images appear and we see the sheer variety of what we have collected.
From tiny particles of the palest white to black rounded beads, we're the first to see what coats Mars and makes the skies pink.
Zooming in, we see finer and finer particles, right to the smallest grains that our optical microscope can make out. We have run out of resolution and that is why we have our atomic force microscope to look even more closely at the dust.
Team member Urs Staufer is impatient, but it will be a week or more before we can move in his instrument and try to image the very smallest of the particles.
The next morning, I drive down from the microscopy house to present our first microscope images at the daily Phoenix press conference. I'm no longer a tourist on Mars, but a scientist. And as a scientist, this is about as good as it gets.
SATURDAY 31 MAY - FIRST LIGHT
It was more than eleven years that Mike Hecht, from Nasa's Jet Propulsion Laboratory (JPL), and I first started thinking about sending a microscope to Mars.
The microscope is a key instrument on the Phoenix lander
Today we have our first chance to get images from the optical microscope of the Phoenix lander as it stands on the northern Martian plains.
Yesterday, we had commanded the sample wheel to bring our test substrates under the gaze of the optical microscope, ordered the LEDs to turn on and illuminate the scene, and asked the microscope to take a series of images.
If it all worked, these would be the highest resolution pictures taken from another planet. Today, the data is making its way back to Earth, first transmitted up to Nasa's Odyssey orbiter circling Mars.
Then, as Earth comes into view, Odyssey relays the data packets to the massive parabolic receivers of the Deep Space Network which then give the packets a final push to where the mission scientists are waiting at the science operations center, the SOC, in Tucson.
We gather round Sanjay's laptop in the downlink room, nervously chatting about all that could go wrong.
Wait and see
During testing for the mission, the images had often come out fuzzy - we found it very difficult to keep precise enough control of the sample wheel.
Meca is now working on the surface of the Red Planet
What should have been crisp pictures had become out-of-focus blurs. Sanjay and Hanna had spent long days on the copy of the microscope station at Imperial, working out the intricacies of how to drive the sample wheel precisely. Meanwhile, the Phoenix microscope itself had been cruising towards Mars.
We are about to find out if they learnt how to drive it.
Data is now flowing back and we immediately see that the spacecraft has sent back all of the pictures we had asked it to take.
Sanjay has prepared computer code that will eagerly pull apart the data packets, reconstruct the pictures, and immediately throw the microscope images up on his screen.
Eleven years shrinks down to the one second it takes to see our images cascade on to the laptop. They're as crisp as any we've seen.
The first image is a test. In truth, there's really not that much to look at - a series of white rectangles covering 1mm by 2mm, with two blurred shadows like a couple of fence posts in one corner.
Phoenix made this footprint-shaped impression on Mars with its scoop
These shadows are the first two beams of the atomic force microscope. With this second microscope we hope to zoom in even further.
The silicon beams are thinner than the thinnest tissue paper, but have survived the trip intact. Urs Staufer, who put together the atomic force microscope, is very happy.
As we look closely, we see the network of fine scratches left in polishing the test substrate. The light from the LEDs, bouncing off the substrate and through the optics of the microscope to its camera, is providing the finest detail ever seen on another planet.
The sun is just hitting the Tucson mountains as we emerge from the SOC and head back to the microscope house.
But we're working Mars shifts and it's dinner time for us. We start up the barbecue under the first light of the Arizonan dawn.
It's certainly worth a toast - we're now ready to take the closest look ever at the dust and soil or even ice that Phoenix digs up from below the surface of Mars.
MONDAY 25 MAY - POSTCARD FROM MARS
The feeling is almost overwhelming as we listen to the live commentary from Nasa's Jet Propulsion Laboratory (JPL) in California as Phoenix descends towards the surface of Mars.
There were big smiles as the team realised the mission was "go"
The radar gives us the height: first we're closing hundreds of metres at a time - surely this is too fast.
But in the last tens of metres the retro-rockets kick in. We slow to a crawl, and Phoenix is given the graceful landing that JPL promised.
The science operations center (SOC) raises a huge cheer that must echo across Tucson. We embrace our colleagues, some of whom we've worked with for over ten years while putting this mission together.
Then I hug my wife and kids - they'll be seeing a lot less of me for the next three months.
The first minute after landing Phoenix tells us all it can before concentrating on fanning out its solar arrays.
The first pictures from Mars were studied closely by the scientists
One critical piece of information - we're tilted by a mere one quarter of a degree. Even before we see any images, we know we have landed on the flattest of plains.
There's now a two-hour wait until the first of these images can be sent back to Earth. Time to share a last meal with our families before we walk over to the secure area of the SOC where we will get started with mission operations.
We've trained so often it first feels like another mission simulation, although we now all have broad smiles on our faces.
As the images start coming in, we all feel the focus of the world's gaze slide from Pasadena, where JPL controlled the landing, to where we're standing in Tucson.
First comes the news we have one of the necessary ingredients for a successful mission - we see the solar arrays are deployed and hear that the batteries are fully charged.
Tom's family joined him in Tucson for the landing
Then we get a picture of one footpad showing a light covering of material kicked up by the retro-rockets. Any fine particles lofted during the landing will provide the first sample to look at with the microscopes on Phoenix.
The room gasps and cheers as Phoenix sends back the panoramic images. The cameras are working superbly - we get more images back than we'd ever managed during operations training.
As the small tilt indicated told us, we've landed on a very flat plain. But the landscape is far from featureless.
We see the surface cracked into polygons, a sign of the forces created by the ice we believe is just below the surface. It will be difficult to control the urge to ask Phoenix to dig down as quickly as possible.
But Phoenix has exceeded our expectations and doesn't deserve to be rushed. So far, it has been the perfect traveller, sending us back a postcard on the day that it arrives.
SUNDAY 25 MAY - WAITING FOR PHOENIX
Phoenix lands today. The Imperial team is now in the US and we're settling in to the Phoenix "microscopy house", our home in Tucson, Arizona, for the three months of the mission.
Foil on the windows helps protect against "Mars lag"
In our case, settling in involves a little more than unpacking our suitcases - we're also taping aluminium foil on the inside of all the bedroom windows.
The neighbours must be feeling rather worried about us, but there's a good reason for what looks like the first signs of collective madness.
We're going to be working on a shift system locked on Mars, not Earth, time. As Mars has a twenty-four-and-a-half hour day we'll be drifting in and out of synchronicity with our neighbours and, half the time, sleeping through the day.
We have experts from Harvard University helping us to overcome "Mars lag" - the continuous battle between our body clocks and a mission schedule working to Mars time.
Some of the tools were developed during the missions of the two rovers still trundling round Mars. The foil will black out our rooms, helping promote melanin production when we need to sleep.
As well as potentially helping us to sleep better, installing the aluminium foil keeps us busy and gives us less time to worry - it's difficult not to mentally replay what Nasa calls the "six minutes of terror" Phoenix will have to endure to get down to the surface of Mars.
But we can't long keep away from the heart of Phoenix activity on Earth - the science operations center, affectionately known as the SOC. Hanna and I take the fifteen-minute drive from the microscopy house to where the mission will unfold after Phoenix lands. But it's now eerily quiet in the SOC.
Empty tables are arrayed in front of a huge screen round around which we will be gathering tomorrow. We'll be anxiously awaiting the first message at 1653 (local time) from Phoenix, telling us it has made it down to the surface in one piece.
A huge screen has been set up in the science operations center
To the side, an Earth-bound version of the lander, which we use for testing, stands quietly in a "stage set" of Mars while its brother is fast closing the last million miles to the planet's surface.
There is a one-minute window directly after landing for Phoenix to send its safe-arrival message. The lander will then concentrate on getting its power sources operational. Two fan-like solar panels will unfurl to capture the weak sunshine of the Martian "Arctic".
In the meantime, the orbiters already circling Mars will send back their accounts of the landing, including the running commentary sent by Phoenix to the spacecraft during its descent.
A copy of Phoenix is standing quietly on the "stage set"
By then, we should know Phoenix's fate, with a second chance for confirmation about an hour after landing. The two previous failed missions to Mars - the 1998 Mars Polar Lander and Beagle 2 in 2004 - took weeks before issuing their death certificates. If the worst happens to Phoenix, we'll know for sure the same day.
Tonight, the Phoenix microscope team will be sharing a last supper before landing. It's perfect weather for a barbecue round the pool at the microscopy house.
Fire and water - just the right ingredients for a final meal before Phoenix reaches Mars.
SUNDAY 18 MAY - A GRACEFUL LANDING?
For the last nine-and-a half-months, the Phoenix spacecraft has been looping around the Sun, closing the distance between Earth and Mars.
While the spacecraft's tasks have been few - checking out and preparing the instruments, and producing very slight adjustments to its course - we have been working hard in London preparing for the 90-day mission.
Extreme Mars challenge: Entry, descent and landing
Using a copy of the microscope station on its way to Mars, the Imperial team - Sanjay Vijendran and Hanna Sykulska and myself - have been testing out the operations in the lab. Our aim has been to get to know the instrument (the Microscopy, Electrochemistry, and Conductivity Analyzer - or Meca) inside out before landing.
On Friday, we packed up our microscope to send to the Science Operations Center in Tucson, Arizona. Its job there will be to help us understand the behaviour of its twin on the surface of Mars.
Now, we can do little more before Phoenix reaches Mars except worry. Before launch last August, my kids scrawled "Go Phoenix!" on the beach at Cape Canaveral, and we enjoyed a perfect pre-dawn launch as the Delta II rocket carrying Phoenix arced over the Atlantic.
It is almost time for them to scribe a new command in the desert sands of Arizona: "Stop Phoenix!"
In a few day's time, early on Monday morning, Phoenix will take just seven minutes to screech to a halt on the icy northern plains of Mars.
It will all be over, one way or another, before the first signal even reaches Earth. No human intervention can overrule the landing sequence on the spacecraft computer that will set in motion first atmospheric entry, then a parachute descent and finally a thruster-controlled landing.
There are high hopes for Phoenix following its launch last year
While this sequence unfolds, we'll be waiting in the Science Operations Center for the signals to come back from Phoenix. Our colleagues and family will be there, including the "Meca babies" born to our instrument team in the years it has taken to prepare for the mission.
Their names, together with those of our colleagues who did not live to see Phoenix launch, form part of an eyetest chart for the microscope that is now hurtling towards Mars at 21,000km/h (13,000mph).
Phoenix will have a welcoming party - three orbiters already at Mars have been re-orientated to communicate with Phoenix before plasma blackout temporarily halts radio transmissions.
One of these, the Mars Express orbiter, will play paparazzo and attempt to snap Phoenix streaking through the atmosphere below on its way in to Mars.
As Phoenix slows down, communication will be re-established and a parachute deployed. As Mars only has 1% of the Earth's atmosphere this parachute doesn't create enough drag to slow us down enough for a soft landing, and so thrusters give us the final braking to ensure we touch down at a mere 8km/h (5mph).
Meca carries an "eyetest chart" with dedications to friends and family
In contrast to the bouncing airbag approach used in the last three missions, this is what the mission's principal investigator, Peter Smith, calls "landing gracefully".
The first sample for our microscope will be collected during the descent - a mixture of atmospheric dust and material kicked up by our landing. But it will be a few days before we get a chance to image this material.
In the meantime, the first images we should be seeing will be of the Phoenix lander and its immediate surroundings.
These photos might not be of huge scientific interest but we'll all be feeling like proud parents in Tucson as we share these first pictures with the rest of the world.
Or we could all be standing dumbstruck as the images fail to appear and we realise the mission most of us have spent over a decade preparing for lies as one more piece of space junk on the cruel surface of Mars.
This page is best viewed in an up-to-date web browser with style sheets (CSS) enabled. While you will be able to view the content of this page in your current browser, you will not be able to get the full visual experience. Please consider upgrading your browser software or enabling style sheets (CSS) if you are able to do so.