Following the Questions

Choosing where to begin

When I arrived at the University of Évora, I found a Physics Department where much of the physics being done was peripheral to the international scientific mainstream. I had to choose someone to supervise my Master's work.

I chose José Carmelo.

What drew me to him was that he was doing science with international visibility in an environment where that was far from easy.

He showed me that the circumstances in which science is done need not define the reach of the science itself.

I suspect that lesson has stayed with me ever since.

Carmelo would later change the course of my scientific life.

He introduced me to António Leite Videira. I later worked with Prof. Videira at the University of Évora. And when the time came for my PhD, Carmelo approached Luís Alcácer and Manuel Assunção about supervising the work.

My PhD therefore had three formal supervisors: António Leite Videira, Luís Alcácer and Manuel Assunção.

Their influence on me was not the same.

Learning how to think

My scientific education came above all from Prof. Videira.

He was a theoretical physicist. Spectroscopy was not his field. But he taught me something more important than a particular technique: how to think about a physical problem.

Years later, rereading the acknowledgements of my PhD thesis, I was struck by what I had chosen to retain from those years.

I wrote about the almost magical force that comes from structuring a chain of logical reasoning, and about the intellectual integrity required to teach in a way that helps form an individual.

Prof. Videira was more than my supervisor.

He was my master.

Neither of us belonged to the scientific community in which my PhD problem was situated. His field lay elsewhere, and I was only beginning to learn lanthanide spectroscopy. Yet he challenged me to understand the problem deeply enough for us to aim at Physical Review B, one of the leading journals in condensed-matter physics.

We eventually did, in 1994.

Perhaps that was my first important lesson in science:

Luís Alcácer had proposed the subject of my PhD: polymer electrolytes doped with rare-earth salts, particularly Eu³⁺. I learned how to prepare the samples in his laboratory in Lisbon, and he gave me considerable freedom to explore the spectroscopy.

Manuel Assunção opened the doors of the laboratory where he worked at the University of Aveiro, making possible measurements I could not perform in Évora.

I could hardly have imagined then that Aveiro would later become my scientific home.

Building bridges

Alcácer later introduced me to Verónica de Zea Bermudez.

She knew how to make organic–inorganic hybrid materials. I brought lanthanide spectroscopy.

But I did not know how to make the materials.

That difference became a bridge.

In 1997, Verónica presented work we had done together on a di-ureasil host. At the same international conference, Sidney Ribeiro independently presented work on the same host.

We met personally about a year later, and a collaboration began. Sidney would become not only a close scientific collaborator, but also a good friend.

Rute Ferreira entered the story during this period as my PhD student. She would later spend six months working in Sidney's laboratory.

A bridge that began around one scientific question was already becoming useful to the next generation.

Questions about host-to-lanthanide ion energy transfer brought Oscar Malta into the journey.

Oscar brought mathematical rigour to the calculation and interpretation of energy transfer from the host network to the lanthanide ions. His deep knowledge of lanthanide spectroscopy repeatedly helped me look more carefully at what the spectra were actually telling us.

Over the years, Oscar would also become a good friend.

This quantitative understanding of energy transfer would later become important in luminescence thermometry, where temperature-dependent energy-transfer processes govern the sensing features of certain luminescent thermometers.

The sample that changed colour

At some point, we began working with hybrid materials containing both Tb³⁺ and Eu³⁺ complexes.

At room temperature, under ultraviolet excitation, the samples emitted predominantly red light from Eu³⁺.

Then one day, a freshly prepared sample came out of the refrigerator emitting green light.

As it warmed, the colour changed.

The explanation lay in the temperature dependence of the energy transfer between the host and the lanthanide ions, and between Tb³⁺ and Eu³⁺.

What had first appeared as an unexpected colour change became a way of measuring temperature through luminescence.

Carlos Brites entered the story at this stage as my PhD student and became central to the development of luminescence thermometry and to the intellectual continuity of the work.

At first we worked at macroscopic scales.

Then came nanoparticles, local temperature measurements, magnetic hyperthermia and nanomedicine.

By 2012, Dani Jaque and I were working on luminescence thermometry in parallel. That year, we both published early reviews on the subject in the same journal.

We could easily have become competitors.

Instead, Dani, his group and ours became close collaborators.

And Dani became a friend.

Some of our best discussions happened while running early in the morning at scientific conferences. The conversations rarely stayed confined to science.

After a while, however, I became less interested in preparing yet another luminescent thermometer.

At some point, the thermometer itself became less interesting to me than the questions it allowed us to ask.

A number that stayed with me

In 2013 we published experiments in which luminescent nanoparticles were dispersed in water inside a capillary heated at one end.

We followed the local temperature response and extracted a characteristic velocity.

At the time, we interpreted it as heat propagation through the nanofluid.

The paper was published.

But the number stayed with me.

I began looking for physical processes that occurred on the same velocity scale.

In August 2013, I met Xiaogang Liu.

I already had the question. His group could prepare much more homogeneous and efficient upconversion nanoparticles.

I convinced him to join the journey.

We redesigned the experiment to control the heat flow and the detection volume more precisely, using those nanoparticles as luminescent probes.

We dispersed them in water and in several organic solvents.

The characteristic velocities remained of the same order as those measured in 2013. But something new appeared: the values depended on particle concentration.

The measured values were compatible in order of magnitude with predicted instantaneous ballistic velocities of Brownian nanoparticles and with an earlier measurement on a single microparticle.

A one-dimensional transient heat-conduction model could not account for them. Thermophoretic velocities were about six orders of magnitude smaller.

In 2016, we associated the measured quantity with the instantaneous ballistic velocity of the Brownian nanoparticles.

It was strong evidence.

Even so, the picture felt incomplete.

After the paper was published, discussions with Dani kept the question alive and helped me think more carefully about what we were actually measuring.

If the interpretation was correct, the instantaneous Brownian velocity should increase with temperature.

That gave us a test.

In exploratory experiments with organic liquids such as ethanol, cyclohexane and chloroform, the measured velocity increased with temperature as expected.

Water was different.

At a characteristic temperature, its behaviour changed.

And suddenly, the solvent became the question.

Water

Water is molecularly simple and collectively extraordinary.

I began reading a literature that was largely new to me: hydrogen-bond networks, anomalous thermodynamic behaviour, competing local structures, structural fluctuations.

And the experiment began to turn itself around.

Instead of using water to study the nanoparticles, perhaps we could use the nanoparticles to study the water.

Experiments and molecular-dynamics simulations suggested a crossover involving local low-density (LD) and high-density (HD) environments associated with different organisations of the hydrogen-bond network.

Around a characteristic temperature, the organisation of these local environments changed, giving rise to the crossover we observed experimentally.

Particularly intriguing was the possibility that, around this crossover, the characteristic spatial extent of these fluctuations could become comparable to the size of the nanoparticles themselves.

Xiaogang called those nanoparticles “nanorulers”.

But subsequent experiments made the ruler less innocent.

The crossover temperature, T_c, was not fixed.

Particle size mattered. pH mattered. Surface properties mattered. Effective surface charge mattered.

At first, this looked like a problem.

If the probe changes what it measures, how can it tell us something about the water?

Then the difficulty itself became interesting.

The thermometer had become a probe, and the probe could no longer be treated as a passive observer.

Different nanoparticles, interfaces and experimental techniques increasingly suggested that what became observable depended not only on the water, but also on the probe through which we were looking.

Testing the interpretation

The idea survived changes of probe and technique.

With the fluorescent protein EGFP, luminescence thermometry yielded Brownian velocities in the same range as those previously reported using a distinct technique, and revealed systematic dependencies on concentration and temperature.

Independent optical-tweezers experiments on individual upconversion nanoparticles, although probing diffusive rather than instantaneous ballistic velocity, have provided compatible evidence on the scale and temperature dependence of the motion.

I would still like a formal proof of the 2016 interpretation.

But the pieces increasingly point in the same direction.

Experiments involving H₂O and D₂O, different probes and different interfaces also changed the way I began to think about T_c.

It could not simply be regarded as an intrinsic thermodynamic temperature of water revealed by a neutral observer.

The interaction between water and probe had become part of the question.

That is where the problem is now.

Following people

Looking back, I no longer see the scientific journey as a sequence of questions separated from the people who helped me pursue them.

Some encounters brought something I lacked. Others changed the way I thought. Some began as teacher and student and became scientific partnerships. Some collaborations became friendships. And some bridges built for one question later made entirely different questions possible.

Rute and Carlos are particularly important to me in this respect.

Both began as my PhD students, at different moments of the journey — Rute with the hybrid materials, Carlos as luminescence thermometry was emerging.

Over the years, both became part of the intellectual continuity of the work: people with whom questions were refined, methods were built, students were formed and new directions became possible.

João Rocha changed another part of me as a scientist.

I served as vice-director of CICECO while he was director, and we later collaborated scientifically, including in luminescence thermometry with metal–organic frameworks.

But his influence went beyond particular projects.

Working closely with João helped me develop something I increasingly came to value: the ability to construct and tell a scientific story — to find the thread that makes a set of results intelligible.

The people I have worked with have also built part of the scientist — and the person — I became.

We build and are built.

Some of those bridges also remain available to others.

Rute once crossed the bridge to Sidney's laboratory.

Years later, Ramon Filho developed the pressure experiments on water in Dani's laboratory in Madrid.

I am also struck by how much of the journey grew out of choices that seemed quite local when I made them: choosing a person to work with, following an unexpected result, entering a field I barely knew, building a bridge because a question had taken me somewhere I could not go alone.

Each choice opened some paths and closed others.

But it also changed the person who would make the next choice — sometimes subtly, sometimes profoundly.

We choose our paths, but the paths we choose also shape the person who will make the next choice.

Still wondering

Following the questions has never meant knowing where they will lead.

Over the years, I have come to know some of my limitations. There are problems I am not equipped to solve, and fields in which other groups are faster, technically stronger or better resourced.

Knowing that helps me choose.

I tend to be drawn to questions slightly to the side of the obvious road — insufficiently explored, sometimes almost incidental, and where moving between different areas may reveal something that disciplinary boundaries hide.

I can be obsessive in pursuing a question.

But I do not become attached to the answer I first imagined.

This has happened repeatedly. A measured velocity became a question about Brownian motion. Brownian motion became a question about water. Water became a question about hydration layers. The apparent limitation of the probe became a question about observability. And the nanoruler may yet become a way of accessing spatial and temporal scales of fluctuations that are otherwise difficult to see.

I do not know where that question will lead.

That is precisely why it interests me.

Discovery, for me, has always had an inner dimension.

The landscape that takes shape within us along the way is often emotionally stronger than the pleasure of arriving at any particular destination.

And occasionally, for a brief moment, I feel again the sweet excitement of the curious child I once was.

Or perhaps I still am.