1.- Los derechos de autor no pueden situarse por encima de los derechos fundamentales de los ciudadanos, como el derecho a la privacidad, a la seguridad, a la presunción de inocencia, a la tutela judicial efectiva y a la libertad de expresión.

2.- La suspensión de derechos fundamentales es y debe seguir siendo competencia exclusiva del poder judicial. Ni un cierre sin sentencia. Este anteproyecto, en contra de lo establecido en el artículo 20.5 de la Constitución, pone en manos de un órgano no judicial -un organismo dependiente del ministerio de Cultura-, la potestad de impedir a los ciudadanos españoles el acceso a cualquier página web.

3.- La nueva legislación creará inseguridad jurídica en todo el sector tecnológico español, perjudicando uno de los pocos campos de desarrollo y futuro de nuestra economía, entorpeciendo la creación de empresas, introduciendo trabas a la libre competencia y ralentizando su proyección internacional.

4.- La nueva legislación propuesta amenaza a los nuevos creadores y entorpece la creación cultural. Con Internet y los sucesivos avances tecnológicos se ha democratizado extraordinariamente la creación y emisión de contenidos de todo tipo, que ya no provienen prevalentemente de las industrias culturales tradicionales, sino de multitud de fuentes diferentes.

5.- Los autores, como todos los trabajadores, tienen derecho a vivir de su trabajo con nuevas ideas creativas, modelos de negocio y actividades asociadas a sus creaciones. Intentar sostener con cambios legislativos a una industria obsoleta que no sabe adaptarse a este nuevo entorno no es ni justo ni realista. Si su modelo de negocio se basaba en el control de las copias de las obras y en Internet no es posible sin vulnerar derechos fundamentales, deberían buscar otro modelo.

6.- Consideramos que las industrias culturales necesitan para sobrevivir alternativas modernas, eficaces, creíbles y asequibles y que se adecuen a los nuevos usos sociales, en lugar de limitaciones tan desproporcionadas como ineficaces para el fin que dicen perseguir.

7.- Internet debe funcionar de forma libre y sin interferencias políticas auspiciadas por sectores que pretenden perpetuar obsoletos modelos de negocio e imposibilitar que el saber humano siga siendo libre.

8.- Exigimos que el Gobierno garantice por ley la neutralidad de la Red en España, ante cualquier presión que pueda producirse, como marco para el desarrollo de una economía sostenible y realista de cara al futuro.

9.- Proponemos una verdadera reforma del derecho de propiedad intelectual orientada a su fin: devolver a la sociedad el conocimiento, promover el dominio público y limitar los abusos de las entidades gestoras.

10.- En democracia las leyes y sus modificaciones deben aprobarse tras el oportuno debate público y habiendo consultado previamente a todas las partes implicadas. No es de recibo que se realicen cambios legislativos que afectan a derechos fundamentales en una ley no orgánica y que versa sobre otra materia.

Este manifiesto, elaborado de forma conjunta por varios autores, es de todos y de ninguno. Si quieres sumarte a él, difúndelo por Internet.

visto en Escolar

Gran Noticia !

Rappin' about CERN's Large Hadron Collider! Links below…

There has been a lot of interest in the original mp3, lyrics, and vocals for remixing. You can find all that here:
msu.edu/~mcalpin9/lhc_rap…

Images came from:
particlephysics.ac.uk, space.com, the Institute of Physics, NASA, Symmetry, and Marvel

The talented dancers doubled as camera people, with some work by Neil Dixon. Stock footage is CERN's.

Will Barras is responsible for the killa beats:
ling.ed.ac.uk/~s9527813/

And thanks to MC Hawking, who first thought of using computer-voice to bring Stephen Hawking to the world of rap :-)
mchawking.com/

The rapper has a day job (we agree this is a good thing) as a science writer.
katemcalpine.com

They'll have a lot of data to sort. 15 million GB per year, actually. Want to get involved and donate your computer's downtime? Try LHC@home:
lhcathome.cern.ch/lhcathome/

Pinche aquí para ver el vídeo

No sabía qué poner para endulzar el post este tan feo, asi que he optado por lo que no falla nunca, Pearl Jam y en este caso concreto, 'All Those Yesterdays' . 

Hasta pronto FaqOff !

Pinche aquí para ver el vídeo

Faqoff no es sólo un blog, es un proyecto que lleva en funcionamiento 3 años y que sumando foro, web y blog ha despachado 4 millones de visitas, 2 de ellas en el último año. Por supuesto que hay cifras muy superiores, pero la mayoría de ellas son o de blogs comerciales que directa o indirectamente viven de esto o de gente que trabaja en temas relacionados con la informática. Lo nuestro era un hobby que afortunadamente o desafortunadamente tuvo algo de éxito y nos exigió una dedicación y responsabilidad que a priori no pensábamos tener.

Esto no es morir de éxito, eso sería echarnos demasiadas flores y por otro lado esto no se muere del todo. Se quedará vivo, pero estático. Enchufado a la máquina, con lo justo para respirar por si algún día necesitamos recuperarlo.

Bueno, si os interesa el tema ya pondre mas teoria xq lo tengo que estudiar para la asignatura de computer security.

El caso es q ya que no he dormido nada esta noche (son las 8 y pico d la maniana XD) me he puesto a armar con multitud de cosas, entre ellas esta. Asi que ahi va la explicacion de como hacerlo en linux. Se puede hacer tambien en windows, supongo que no sera muy diferente, pero yo no lo he hecho:

La pagina del proyecto:

http://www.torproject.org/

Primero de todo, instalar los dos programas que vamos a necesitar. El TOR en si y un proxy que mande todas las conexiones a traves del TOR.

    $ sudo aptitude tor
    $ sudo aptitude privoxy

Ahora hay que configurar privoxy para que mande todo a traves de tor, para eso:

    $ sudo vim /etc/privoxy/config

id al apartado 5, donde pone forwarding y en el segundo punto (no el http, el que pone socks4) poned estas 4 reglas:

    forward-socks4a / 127.0.0.1:9050 .
    forward 192.168.*.*/ .
    forward 10.*.*.*/ .
    forward 127.*.*.*/ .

la primera es para enviar todas las conexiones al puerto en el que el tor esta escuchando
las otras tres es para que no se pueda acceder a vuestra red local/ordenador desde la red tor.

Eso es todo lo que hay que configurar de esos programas, de todas formas, si quereis darle un repaso hay mas opciones ahi, el archivo para configurar el tor es

/etc/tor/torrc

Una vez hecho esto reiniciad los demonios (que se activan automaticamente al instalar los programas):

    $ sudo /etc/init.d/tor restart
    $ sudo /etc/init.d/privoxy restart

Vale, ahora para usar la red lo que teneis que hacer mandar las conexiones al puerto 8118 de vuestro propio ordenador (localhost). Esto estara conectando con privoxy, que esta escuchando ahi. A su vez el enviara todo a traves de TOR. Este a su vez ira enviando entre servidores y finalmente enviara al servidor de destino. Hay que tener en cuenta que la ultima conexion de todas, la que va desde el ultimo servidor de la red tor hasta el servidor que querais, no esta encriptada. A efectos practicos: da igual.

Bien, para usar todo este invento con HTTP lo mas facil es usar firefox. Lo primero, instalar estos 2 plugins:

•     https://addons.mozilla.org/en-US/firefox/addon/2275

pone un boton en la esquina inferior derecha del firefox, click activa, click desactiva. configurarlo: marcad la casilla para que utilice privoxy en las conexiones

•   https://addons.mozilla.org/en-US/firefox/addon/82

plugin para controlar las cookies de diferentes sitios

Una vez hecho eso, activad el torbutton e id a la web:

https://torcheck.xenobite.eu/

Si sale algo como:

    Your IP is NOT identified to be a Tor-EXIT.
    So you are NOT using Tor to reach the web!

en rojo chillon es que algo esta mal, repasad el proceso. Podeis ayudaros con:
https://wiki.torproject.org/noreply/TheOnionRouter/TorFAQ#head-417dc5037b9cd0f51ec305a2a8ec78e7348cf7f6

que es la faq de tor sobre como usar el cliente tor.

Si os sale:

    Your IP is identified to be a Tor-EXIT.
    So you are using Tor successfully to reach the web!

en verde bonito, pasad de las siguientes lineas porque ya os funciona. Click en "START" para ver que cosas os pueden delatar y hacedle caso si quereis, depende de vuestro nivel de paranoia.

Posibles sitios de confusion:

• Configuracion del privoxy para enviar las cosas al tor. Aqui es donde yo he estado atascado un rato, pero con copiar las mismas reglas que tengo yo deberia funcionar. Si estais detras de un firewall o alguna cosa rara teneis que abrir ciertos puertos, viene todo especificado en el faq

• No estais enviando nada al privoxy: mirad a ver si teneis bien configurado el torbutton.Para ver que efectivamente estais usando el tor los logs estan en:

/var/log/tor/
/bar/log/privoxy/

Si quereis verlo en tiempo real podeis parar los dos demonios y ejecutarlos a mano para que vayan poniendo en consola lo que va pasando:

    $ tor

y en otra consola

    $ privoxy /etc/privoxy/config –no-daemon

Cuando lo hayais arreglado, ctrl+c en las dos consolas y volveis a iniciar los demonios con:

    $ sudo /etc/init.d/tor start
    $ sudo /etc/init.d/privoxy start

Listo !!!

[tags]Linux, seguridad, TOR[/tags] 

Se conecta la clavija Mini-Jack al cable Jack que estará conectado a la guitarra. El extremo del Mini-Jack se conecta a la tarjeta de sonido, exactamente en la Entrada de Linea [en mi caso azul]. Abrimos nuestro controlador de sonido y activamos Entrada de Linea y subimos el volumen bien alto.

Ahora mismo ya podemos tocar y escuchar por los altavoces nuestra guitarra. Para grabar, lo hacemos mismamente con el Grabador de Sonido que trae Ubuntu. Abrimos el programa, Aplicaciones > Sonido y Video > Grabador de Sonido. Elegimos el formato que queremos para nuestro futuro archivo de sonido. Importante, tenemos que seleccionar que el programa coja el sonido de la Entrada de Linea. 

Una vez hecho, solo queda grabarse haciendo el indio o despertar a los vecinos con canciones mutiladas por nosotros mismos.

Elderly Woman Behind a Counter in a Small Town 

Y ese es un ejemplo de como suena. Soy malo, lo sé, pero sólo quería probarlo.

+Info sobre guitarras en Linux 

Actualización :

Podemos instalar algunos paquetes para tener distorsión y efectos en el ordenador.

$ sudo aptitude install jackd qjackctl jack-tools creox 

• Jackd – servidor de audio – nos sirve para tomar la entrada de la tarjeta de sonido
• Qjackctl – GUI para controlar jackd
• Creox – preprocesador de audio – con esto haremos los efectos de la pedalera

Seguid esta guía que está bastante clara. [+] 

[tags]Ubuntu, guitarra, jack, grabar[/tags]

Nota : El autor de este post, inmotra, está en New Orleans y su ordenador ha sido comprado en EEUU y no tiene ñ. Tampoco tiene acentos, asi que perdonad las abundantes patadas al diccionario.

estoy siguiendo la web:

http://www.howtoforge.com/ubuntu_dm_crypt_luks
y
http://ubuntu-tutorials.com/2007/08/17/7-steps-to-an-encrypted-partition-local-or-removable-disk/

en esta ultima viene todo perfectamente en 7 pasos

venga, ahi va la explicacion de como hacerlo:

Primero hacer una copia de seguridad de todo lo que haya dentro con cp, eso antes que nada

luego instalar las herramientas necesarias y cargar los modulos del kernel:

$ sudo aptitude install cryptsetup hashalot

para poner los modulos:

$ sudo modprobe aes
$ sudo modprobe dm-crypt
$ sudo modprobe dm-mod

y meterlos en /etc/modules para que se carguen siempre (a no ser que seas un paranoico del copon y no quieras que esten cargados por alguna extrania razon… que entonces es mejor cargarlos cada vez que los vas a usar):

$ sudo vim /etc/modules

vete al final y aniades esos 3, es decir, 3 lineas que sean:

aes
dm-crypt
dm-mod

cuidado con ese ultimo xq en el link que he puesto el dm-mod viene con guion bajo pero es un guion normaldesmontar el usb:sudo umount /dev/sdb (en mi caso, en el tuyo donde este puesto, si es q esta puesto)borrar la particion y crear dossudo fdiskp   (para imprimir, si hay una sola particion entonces mira el numero y borra esa, si hay varias probablemente sepas el xq y puedas elegir la que quieras quitar)d<num>    (el numero de la particion que quisieras borrar)

n
p
1

   a la primera pregunta no respondas, a la segunda ponle +numMB, donde num es el numero de megas de la particion (1000 es un giga). esta sera la particion "publica" sin encriptar, el espacio restante sera el de la encriptada

n
p
2

enter, enter , a las dos preguntas

ten en cuenta que la segunda particion tiene que tener por lo menos 256 sectores, sino no te va a funcionar el invento

ahora a borrar todos los datos que hubiera todavia en el disco duro para que no puedan recuperarse mas que con microscopios electronicos y metodos estadisticos para ver que bit era cada cosa que tienen en los de la NSA (eso segun nuestro amadisimo profesor de seguridad)… pero vamos, para nuestro proposito creo que es suficiente, espero no estar ayudando a ningun terrorista a guardar los planes de nada en su usb..
para hacer eso:

$ sudo dd if=/dev/random of=/dev/sdb2 (en mi caso sdb2, pero va cambiando si quitas el usb y lo vuelves a poner, asi que hay q mirarlo bien a la hora de hacerlo)

cuidadin con no confundirse de particion que si borras algo asi no lo recuperas ni pagando todo lo que puedas pagar…
por cierto que tarda un bueeeeeeeeeeeeeeeeeeeen rato en hacerlo. unas cuantas horas normalmente, depende de lo grande que sea la particion que quereis limpiar. yo personalmente me aburri despues de hora y algo y solo habia grabado 16Kb… asi que va para muuuuy largo. casi es mejor hacer toda la movida y luego meter el archivo mas grande que entre y borrarlo despues, como esta encriptado el resultado sera informacion "aleatoria" no susceptible a ataques estadisticos o de cualquier otro tipo (vamos, eso si nos fiamos de que el algoritmo usado por las herramientas estas sea bueno, que hemos de suponer que si)

lo que hace el dd es llenar la particion con la informacion que salga de /dev/random, que es el generador de numeros aleatorios de linux. asi estara poniendo 0 y 1's a discreccion sin ningun orden, dejando ilegible lo que habia debajo. cuando se borra un fichero lo unico que se hace es dejar de referenciarlo en el disco, con lo cual no lo puedes encontrar, pero sigue estando ahi. alguien que sepa como buscarlo lo puede llegar a encontrar facilmente

esto es solo para la primera vez, es para darle formato a la particion:

$ sudo partprobe        (para que el kernel reconozca las nuevas particiones)

$ sudo cryptsetup luksFormat /dev/sdb2 -c aes -s 256 -h sha256
$ sudo cryptsetup luksOpen /dev/sdb2 CRYPTO
(vosotros el que sea, dadle al tabular dos veces para saber cuales son los posibles y poned el que os toque)
$ sudo mke2fs -j /dev/mapper/CRYPTO -L CRYPTO

y ya que estamos, formateamos tambien la otra con fat (para poder leerlo desde windows), no es un formato que a mi me encante, pero es el que menos problemas da a la hora de usar el USB en cualquier sitio:

$ sudo mkfs.vfat /dev/sdb1  (si es que habeis seguido las instrucciones que os he puesto, elegid en vez de la "b" la que os toque a vosotros)

vale, ya esta todo listo, ahora lo mejor sera que lo desmontes todo y uses los scripts de mas abajo para montar/desmontar. coge el de desmontar y ejecutalo. luego el de montar, ejecutalo y listo, ya lo tienes funcionando.

Script para montar y otro para desmontar:

$ sudo gedit mount.sh    (yo no uso gedit, pero lo mismo os gusta mas, personalmente lo he hecho con mi amado vim)

#!/bin/sh
sudo cryptsetup luksOpen $1 CRYPTO
sudo mkdir /media/CRYPTO
sudo mount /dev/mapper/CRYPTO /media/CRYPTO

para ejecutarlo:

$ sudo sh mount.sh /dev/sdb2

vale, ahora prestad atencion. no siempre es sdb2, va cambiando cada vez que enchufas el USB al ordenador, asi que unas veces puede ser sdb2, otras sde2, sdf2…o lo que sea. para saber cual es escribid el comando y /dev/sd entonces dadle dos veces al tabulador y saldran las posibilidades. pues descartad las que conozcais y quedaos con las que sepais que es (si veis una que tiene mas de 2 particiones obviamente no va a ser

para desmontarlo usad este script:

$ sudo vim umount.sh

#!/bin/sh
sudo umount /media/CRYPTO
sudo cryptsetup luksClose CRYPTO
sudo rmdir /media/CRYPTO

para ejecutarlo:

$ sudo sh umount.sh

y listo !!

cuando enchufeis el USB al ordenata dadle a cancelar al cuadro de dialogo que puede que os salga diciendo: este volumen esta encriptado, introducir la clave (o algo por el estilo), si poneis la clave ahi hara como que todo va bien, pero en realidad no lo montara bien (vamos, a mi no me lo monta bien) y parecera que el USB esta vacio… asi que no hacer eso

usad los 2 scripts estos y todo ira como la seda.

para copiar cosas:

$ sudo cp loquesea /media/CRYPTO

se necesitan permisos de root para meter cosas. se pueden cambiar los permisos de /media/CRYPTO  para que solo root pueda leer o moverse por los directorios/ejecutar cosas. si haceis esto ya es lo mas seguro que podeis tenerlo, xq necesitareis primero la clave del USB y permisos de root para montarlo y luego permisos de root para leer/escribir, con lo cual aunque dejeis el ordenador en una sala llena de gente nadie os podra robar las fotos de vuestra novia/novio/madre/loquesea

Advertencias:

para encriptar y desencriptar se necesita tiempo, asi que cuidadin con los archivos que meteis. podeis tiraros un rato laaargo laaargo para meter algo de unos cuantos gigas… pero si lo haceis esperad, xq aunque se quede como pegado si que lo estara copiando

si usais esto para almacenar cosas ilegales yo no quiero saber nada, ni os he dado ideas ni nada. esto es solo para armar y aprender a hacer cosas interesantes.

si estais en estados unidos cuidadin que lo de encriptar cosas a veces esta un poco mal visto y puede que los del gobierno, si sospechan que teneis movidas raras ahi dentro, se enfaden un poco jejeje

y recordad que todo esto depende de la clave que pongais, asi que usad una de por lo menos 8-10 caracteres alfanumericos y simbolos, mi clave es:

jajajajajajajajajaja, pero estamos tontos ?

[tags]Linux, Ubuntu, USB, encriptar, seguridad, tecnología[/tags] 

The polymerases are the enzimes used for DNA replication. They can be helped by “clamps” that improve the processivity increasing it 500,000 times. Ribosomes themselves are clamps that closes around the RNA message.

There are a lot of other enzimes that modify and interact with the nucleic acids, like:

• polymerases

• nucleases: cut acid strands.

• Ligases: connect nucleic acid strands.

• Repressors, transcription factors, enhancer proteins and another regulatory proteins: regulate the use of the information

• Base-excision nucleases: remove bases from nucleic acid strands.

• Topoisomerases: solve the topological problems of long strands of DNA.

• Recombinases: swap DNA portions of different DNA strands.

• Spliceosomes: edit RNA removing pieces

• Nucleosomes and other proteins: package nucleic acids for storage.

All this tools are helpful for editing nucleic acids to write the blueprint for custom proteins.

There is an intermediate step between the DNA and the proteins. DNA first is coded in messengerRNA (mRNA) and then ribosomes produce proteins from that information. There are some virus that skip that step, but that is not usual. That can be an evolutionary legacy.

That step is called transcription. Each DNA base correspond to a RNA base. RNA polymerases display the DNA strand and connect the RNA bases.

Then the ribosomes translate that RNA into proteins. That is a translation from one language to another. That translation is done by the transfer RNA (done itself by RNA). The genetic code is based on sequential triplets. Each 3 nucleotides form a codon and specify one of 20 possible amino acids. There is also some of them that codes the “start” and “stop” states.

The ribosome is a complex machine. It gets the codons, one of a time, and assemble the associated amino acid into the protein. Then moves to the next codon and repeat the process. It can assembly around 1000 amino acids per second. This extraordinary machine is composed primarily of RNA.

All this information is stored in 4 different bases, what is 2 bits. About 30 atoms are needed per base, so 15 bases are needed for store 1 bit. But the retrieval of that information is 1million times slower than in modern computers.

Many reactions in nano scale require energy, it can come from electrical, light or chemic reactions. In cells the energy used is not as in the macroscopic world, where we waste a lot of it, instead the energy used is the required, because all the heat is rapidly dissipated so it isn't useful (although it can be useful to raise the temperature of the organism). The reactions inside the cell work together to be possible. One process can be not favorable, but with another one can change, so both together are possible.

In cells the energy is stored in molecules. Those molecules are instable and provides the energy needed in many reactions. They are the fuel of the organisms. The most common is the ATP. It is created with energy from the light or from the breakdown of the food. Then release that energy in unfavorable biomolecular processes.

The ultimate energy source of almost all life is the Sun. The light is captured by small molecules called photosynthetic reaction centers. They get a photon and create a high-energy electron which is used for power. This electron is properly stored to avoid its energy lose and then is placed on a carrier molecule. The hole then is filled by another low-energetic electron usually provided by water.

Photosynthetic organisms absorb a lot of different wavelengths using chlorophyll and carotenoid molecules.

The charge transport is important in macroscopic world because it allows multiple machines to work. Cells also can conduct electrons, but instead of do it on bulk thanks to a potential difference do it electron by electron. For this purpose there are carriers that tunnel the electrons in distances between them less than 1.4 nm. If the total distance needed is more than that chains of carriers are used. This is used for two main purposes: the bulk delivery of electrons for chemical reactions and for powering other processes. Although this is possible there is no use of the single-electron computation. The biological computation is performed by hard-wired and biochemical networks and at the micro scale by programmable nerve networks.

DNA can conduct, but it is really difficult to experiment with it because getting a single strand of DNA, isolate it and apply two electrodes is a major challenge. It is supposed to conduct about 10E+12 electrons per second, not bad for a single molecule of DNA. But this is not used for anything in nature.

It is possible to store energy in a difference of charges. Ions instead of electrons. Membranes impermeable to ions can act as the medium between the different charges. The ions are pumped across the membrane to the other side and then the flow back is used to perform chemical or mechanical work.

Chemical transformations are done in cells with the help of catalysts, the enzimes. They minimize the necessary energy of the process and speed it controlling the side products. In the active site of the enzyme specific amino acids are placed to stabilize the transition state of the molecule undergoing.

To speed up the reactions, enzimes reduce the entropy. It is done in the active site, which is separated in two regions, a specificity pocket that recognizes the proper substance and binds to it and the catalytic machinery that performs the chemical transformation. They can be very specific, separating molecules with differences in a single atom or stereoscopic differences. They also stabilize transition states modifying the target to make it more active.

There are some possibilities to control the regulation of the processes. We already talked about the power control using ATP, but there are another mechanisms.

Allosteric motions are used for regulation. A protein can adopt two (sometimes more) possible states: one relaxed (R state) when the protein binds tightly to substrate molecules, and another tense (T state) when the protein resists the binding. We can force the protein to adopt one of these states, turning it on or off. It is also possible cooperative binding, so when binding to one substrate the protein increases the affinity for binding additional molecules at the remaining sites.

There are two different sites in an allosteric complex, one for the substrate and another one for binding a regulatory molecule. Binding to this second site changes the shape of the other site, changing the affinity of the substrate binding site. The control depends on the linking of the two sites within the protein complex. This mechanism provides a vast number of possibilities for regulation.

The action of biomolecules can be also changed by changing the groups bonded to them. If positive or negative charges are added to them the key chemical groups can be altered. So it can be used to activate or deactivate the function of the proteins.

Biomaterials are not as the materials of the macroscopic world, they are not constructed to last for a long time. Instead they are always being constructed and destructed to construct another things to give a dynamic response to the environment.

Filaments are an important biomaterial. They are created by binding a protein with different copies of itself forming structures. They can form linear filaments, microtubules or other structures. It depends of this conformation the rigidity or flexibility of the structures.

Some other bigger structures are possible. They can be built with subunits combined. They can form networks strong but resilient at the same time, porous and permeable to water and small molecules. The most common is a though two-dimensional membrane to cover the lipid membranes. They can also form a network of proteins inside a cell or create a three-dimensional network used for support and transportation.

Some minerals can be combined with biomaterials to get additional strength or another properties like sense the gravity, the magnetic field of Earth or for vision. Biomineralization is the process of growing crystals on demand. There are some methods, but all of them follow the same patron: get a space to grow the crystals, transport ions inside, nucleate crystals or aggregates of the mineral and finally control the growth and orientation of the mineral.

This use of organic-inorganic materials is very important for bionanotechnology because it is possible to combine the strength with the resilience. It is still being studied, but now it is better understood, although is still a challenge.

There are elastic proteins that are formed by disordered chains that may be stretched and distorted. They have multiple uses in organisms.

Another biomaterial is adhesives. They have to follow two criteria: stick to the surfaces and form a tough solid to be stable themselves. They can be compromised by water, because it can solve them.

Motors are rare in organisms, but there are a few. They are used for several tasks, like: separate the chromosomes or the remodeling of cell organelles. To the first one the two best exampled studied are myosin and kinesin. They use ATP to energize the process.

Another motors are the rotary motors, like in flagellar movement. These motors use also ATP to provide the energy necessary.

The brownian ratchets are motors that use the thermal motion. They use a barrier that allow protons to go pass in one direction, but not in the other one.

All the bionanomachines inside the cell are freely diving, so it is necessary a container. That's the membrane. They are impermeable to those molecules but permeable to others. A complete sealed membrane would be useless for a cell. To provide that permeability the membrane use channels that are passive transport devices. They allow the flow of molecules through membranes. They can work with a specific molecule and can be gated, closing or opening in response to some signal.

To transport proteins an active approach is needed. Usually a molecule is binded to the protein to force the molecule across in the process. In some cases ATP is used, but light can also be used. The most common types of transport are the ABC transporters. The name refers to ATP-binding cassette and use a flip-flop mechanism. They are like boxes open to the outside of the cell. A molecule goes inside and then cross to the cell.

For sensing the environment receptor proteins are used. They change the shape or change the charges distribution. Then those changes are amplified and are transformed into intracellular changes or initiate nerve impulses.

Bionanomachines are particularly good recognizing taste and smell. In order to accomplish that goal they can recognize specific molecules using proteins. Mammals have about 1000 genes to encode odorant receptors that can be combined to sense billions of different odors.

Light is sensed by monitoring light-sensitive motions in retinal. They change when absorb a photon.

Mechanosensors are still being studied. They can detect touch, acceleration and sound. They are thought to be ion channels that open quickly to allow many ions to pass. They are adaptable so forces like gravity are ignored and only changes in forces produce signals. This is done by relaxing the channel progressively.

Bacteria use a temporal sensing mechanism. They are too small to sense changes in the gradient of nutrients from one side to the other, so they are sensing the concentration and comparing it with the previous data. Their flagellar motor is working to impulse the cell in one direction, if the data says there is less concentration then it reverses and the cell tumbles, picking a new direction.

Self-replication is needed to create macroscopic objects with nanomachines. There are too many molecules in a gram, so a machine that can replicate itself is the best because in a sort period of time many nanomachines can be working. This is a good promise but also it is dangerous.

Cells are self-replicators, the parts include:

• Information-driven assembler: they construct new molecules using the ribosomes.

• Information storage medium: DNA is the storage of the information for the ribosomes.

• Duplicator: DNA polymerase duplicate the information storage in modern cells.

• Chemical processors: They convert available raw materials into building blocks. There are thousands of these enzymes to perform these transformations.

• Infrastructure: They support the cell but also allows transportation inside it. The most famous is the lipid membrane.

It is thought that the most possible living organism (in laboratory conditions) has between 250 and 350 genes. The simplest living organism has 550 but lives inside other cells and uses many of their processes. But not only genes are enough, the blueprint is important but also a map of the structure of a living cell is also important.

This approach is like this because of evolution, but if we create nanomachines without the competition factor then we can make them without some of the parts, like the containers or other characteristics and we can make them more efficient.

In bionanothechnology machine-phase matter is used, it is the combination of multiple nanomachines to perform a task. They are composed of many modular machines, each very stable and functionally efficient. They are redundant, with many copies of the type of nanomachine.

Muscle sarcomeres are one example of machine-phase matter. They combine the action of many myosin to create macroscopic motion.

Neurons are programmable electrical components. They have an input layer (cell body and associated dentrites), an output layer (terminal branches) and a high-speed electrical communication (axons). They can act in an analog to digital schema. They encode the magnitude of the signal into a frequency of action potentials.

Chemists have studied the properties of these atoms and biologists the details of the nanomachines constructed by them. So we have now the knowledge to start creating these new nanomachines. We are going to review the different methods followed to do that.

The natural bionanomachinery is designed for a specific environment. It is done for working inside cells, in a water environment. Without that conditions they don't work properly or don't work at all. The temperature is also important, the most common is 37º. In this environment nanomachines are stable, but they can be destroyed to create new ones with a little energy cost. The life of a nanomachine is between seconds to a year (but this is rare).

The intuitive approach to construct nanomachines is to build them atom by atom. However it is possible to follow a hierarchical strategy. George Whitesides set four different strategies:

- Sequential covalent synthesis: atoms are bonded into covalent molecules. The atoms are placed together piece by piece to build up the structure. The main advantage is the diversity achieved because it is possible to combine in almost any combination.

- Covalent polymerization: some modular units are linked into linear or branched chains. Using this method it is possible to get very large chains. The DNA synthesis is an example of this. There are limitations, for example, once the scheme is chosen it has to be always followed. Another limitation is that only stable chains under reaction conditions can be used to link each other. Enzimes allow the use of many different monomers, it is being studied.

- Self-organizing synthesis: nanostructures are formed by noncovalent bonds from modular units. They bond each other adopting the thermodynamic minimum.

- Self-assembly: as the author says "the spontaneous assembly of molecules into structured, stable, noncovalently joined aggregates". It is the most important in the construction of nanomachines and the less intuitive because it is far from the macroscopic behavior.

The key of bionanotechnology is Carbon. Every organic molecule has carbon. There are other atoms like oxygen and hydrogen also present that bonds to carbon. There are three forces that assure the stability: covalent bonds, nonbonded forces within molecules and interaction with water.

Covalent bonds are the most strong bonds. They are the product of two atoms sharing electrons of their last level. They are expensive in energy to break. Atoms with covalent bonds form the skeleton of the biomolecules. There are a few general rules that control them, so understanding them it is possible to figure out how the molecules are. Single bonds allow rotation, while double or triple don't. The most stable structures are those that are made by carbon bonded. Then nitrogen, oxygen and phosphorus can be bonded. Hydrogen is always present everywhere. The different quantities and order of this elements forms the molecules.

But these rules sometimes are a little bit different. Resonance is a neutral field between double and single covalent bonds. In a ring formed by carbon all the links are equal, no single links and no double links. This allow flexibility in the molecule.

The dispersion and repulsion forces between atoms are powerful. At a close range atoms experiment a dispersion force, but in a closer range a repulsion force begin to work.

Hydrogen bonds are weaker than covalent, but they contribute to stabilize molecules. This bonds are formed between a hydrogen atom covalent bonded to a nitrogen, oxygen or sulfur and another oxygen, nitrogen or sulfur atom. Because of the water environment this is very important for the stability of bionanomachines

.Electrostatic interactions are also active at atom level. They aren't directional and has a wide range. They help to the stability of molecules. These interactions are reduced by the dielectric effect.

Another effect is the hydrophobic. It was explained in chapter 2 and it is still the same. It helps molecules to form aggregates.

The proteins have a structure that provide them the characteristics they have. All the information about the amino-acid sequence and the structure is coded in the DNA. It is really hard to predict the structure from the sequence, but once it will be done custom proteins will be constructed much easily.

But proteins are not always stable. There are many ways for a chain to be stable. Similar proteins tend to have similar structures. That is a really good feature for bionanotechnology because then it is possible to use comparing tables to determine structures. The only but is that a single change where in a wrong place would be fatal for the protein and not match at all the other model.

Proteins have a hierarchical structure that provides them the stability they need. They have some particular structures like the a-helices and b-sheets. This structures provides many hydrogen bonds inside the molecule. This local structures then fold into a stable globular structure.

Those proteins can form a globular structure with a hydrophobic core, this is the positive design. The interaction with water provide most of the energetic stabilization for them. The unfavorable reduction of entropy must be compensated by favorable interactions in the folded structure.

The negative design is to ensure that a single folded conformation is created. The protein is designed to avoid any conformation energetically unfavorable.

To achieve this folding there is a collection of molecules called chaperones that assist proteins. They separate the molecule from the water environment and provide the conditions needed. They are like a box, the protein enter inside and the top is closed. Cells have two chaperones, the first is the formation of disulfide bonds that cross-link cysteine amino acids at distant parts of the polypeptide chain.

There are proteins that can be stable at high temperatures. In nature they are in some bacteria, so it is easier for us to study them and understand how to reproduce it. They are very useful for industry because they are processes where enzimes are needed and they are at high temperatures. The structure of these proteins is almost the same as the heat-labile proteins. The amino acid sequence is not quite different. The major difference is the rigidity. They have in the surface some new ion pair interactions between amino acids, new disulfide linkages or metal ions, incorporation of rigid proline or replacement of flexible glycine amino acids.

Some proteins don't use a static structure, instead they change between structures. This characteristic is attractive in signaling proteins because they can change rapidly to response to some signal. The disorder also allows the protein to have several capacities.

Self-assembly is how most of bionanomachines are build. They associate each other spontaneously in flexible chains that later form compact structures and then into functional complexes. The instructions used are in DNA, but it only specifies the amino acid chain, then they form their own structures. But it is more restrictive than directed construction, where we can specify exactly everything.

This process is modular. Large assemblies may be created with many identical pieces. It requires specific geometry of interaction and unique interaction between subunits to avoid any crosstalk. This can be a serious problem when interaction surface is small because they can interact with another unwanted molecules.

The process is also spontaneous, it doesn't require any guide or additional information. That involves a careful trade-off enthalpy and entropy.

Symmetry is a good help for constructing proteins because it is only needed part of the information to construct the whole thing. It is also good to control the errors. It has also many functional advantages, they can cooperate with their neighboring. And they can use many identical bonding sites to enhance the strength of binding to their target.

There are 3 important classes of geometry: cyclic, dihedral and cubic:

• Cyclic:only one single axis of symmetry forming a ring of symmetrically arranged subunits. Molecules with the higher cyclic groups are used in specialized functions like interaction with membranes or rotational motion.

• Dihedral: a central axis of twofold or higher rotational symmetry perpendicular to another axis of twofold symmetry. They have multiple surfaces of interaction, each different. They are used to construct enzimes that modify their action.

• Cubic: an axis of rotational symmetry with a nonperpendicular threefold axis. There are 3 possible arrangements: tetrahedral, octahedral and icosahedral. Transactional symmetries can be used to extend structures.

Line symmetries: include a transaction in one dimension, adding a rotational symmetry around the transaction axis yields a helix.

Plane symmetries are formed when translational symmetries are applied in two dimensions.

Space group symmetries are rare in nature.

The assembly process can be controlled by enzimes that speed up or stop it.

When it is needed to construct a very large protein the quasisymmetry can be used. It doesn't require as much information as a perfect symmetry. To do that two or more identical molecules are placed at each symmetrical position. This requires that subunits adopt slightly different conformations in the different nonsymmetrical positions.

The self-assembly is promoted by crowded conditions. If there are a lot of molecules in a solution they will tend to interact with each other more frequently than if they are alone. That is why nanomachines work better in crowded conditions, although the larger aggregates are harder to form.

Sometimes a less concrete building material is needed, so self-organization is a perfect method for creating structures that are flexible, resilient and self-repairing. It hasn't the control present in self-assembly, but that's exactly what is needed in some applications. These systems are also modular but don't have specific surfaces of interaction.

Lipids self-organize into bilayers. That is due to the hydrophobic and hydrophilic effect. Each lipid has a distinctive critical concentration and it is very low and lower for lipids with longer carbon chains. They associate to shield the hydrophobic segments of the molecule.

The lipid bilayers are fluid because they are composed of many nonbonded molecules. The lateral motion is fast, but flipping lipids from one surface to the other is not frequent. The fluidity is useful because it allows spontaneous healing of damage. This fluidity is dependent on the structure of the component lipids and on the temperature. They are highly flexible, allowing complex shape transformations. This bilayers can also have proteins to change the permeability or provide some different properties.

Molecular recognition is needed to control the interaction between parts in the bionanomachines.

There are two design concepts postulated by H.R.Crane:

“For a high degree of specificity the contact of combining spots on the two particles must be multiple and weak.”

“One particle must have a geometrical arrangement which is complementary to the arrangement on the other”.

To recognize molecules the nanomachines use some specific spots where the desired molecule, and only it, can be attached. Inside those spots there are a lot of weak interactions between the atoms of both molecules.

It is impossible to achieve high precision in the interaction between molecules because the atoms are spheric and they are discrete, so we can't break them in smaller pieces. Surfaces of interaction are rough, they can't fit exactly.

Because nanotehnology is based on macroscopic engineering the same approach is being taken. So there are rigid components binded by a few mobile bonds or joints. The natural bionanomachines are not like that. Evolution do it the other way around, it take flexible molecules and select some rigid ones for special tasks.

The flexibility in biomolecules is at all levels. From atoms to the higher structure. It is used to enhance their functions. Although many proteins are composed of several rigid domains connected by flexible linkers. Multichain assemblies may shift between specific conformations with different properties. In enzimes this is called “allosteric” motion and is used for regulation.

Biomolecules may incorporate specific levels of rigidity to improve the entropy.

These all kinds of flexibility create a greater level of complexity at design-time. It is harder to predict the structure and properties of the molecules we are trying to construct. But biomolecular flexibility will provide one of the greatest challenges, and potential benefits, of bionanotechnology.

Recombinant DNA technology is the main process to construct this proteins. It is about changing the DNA that stores the information to create the protein and then wait until ribosoms create it for us. We use two natural enzimes for this, restriction enzimes and DNA ligase. Using this we can cut and paste DNA to make our own sequence.Some of this enzimes are being sold in the market, so now it is easy and not so expensive to manipulate DNA. There is a variety of natural biomolecules for handling DNA, like:Restriction enzimes: isolated from bacteria. Used to cut DNA.DNA ligase: reconnects broken DNA strands.DNA polymerase creates a new DNA strand by using another strand as a template.
Once we have a new DNA we can duplicate it using: DNA cloning and polymerase chain reaction.
The DNA cloning create identical chains. To do this we can inject DNA in a virus and let it to inject the DNA in some bacteria. Then they will duplicate, so we will have a lot of identical cells.
The polymerase chain reaction is for coping a small sample of DNA. We use another bacteria for this. Forte it to duplicate. Using this method we can have as many DNA as we need.

Once we have all this DNA we can create the proteins byr forcing the cells to produce it. This is done using expression vectors. They have a highly active promoter sequence that forces the cell to create mRNA. It is taken from a virus. This mRNA will be later processed to create the proteins. The cells used are bacteria. But they have some limitations, animals and plants modify their proteins after create them, but bacteria don't. This can be a serious problem because the immune system can react against them because of those differences. Another problem is that proteins tend to aggregate when they reach high concentrations forming dense inclusion bodies.Proteins can be created as well without the help of living help using a test tube and enzimes that make them. But the synthesis of protein from the mRNA is still a challenge. Stracts of cells can be used then to form those proteins. But  because of the complexity of the system it is only used for small tasks and for research. This is a controlled method and has not the interferences of other enzimes in the bacteria, that's why it is good for research.In other cases we prefer to make some little changes to a natural protein. Then we can use the site-directed mutagenesis. This is to introduce some specific mutations in an existing DNA sequence to create already existing proteins but with some changes. This method has revolutionized molecular biology. It is useful to determine the function of specific aminoacids. It is also used to improve the stability  of proteins.

We can also combine two different proteins to get a new functionality with the older ones. We can specify the attach point for this task.

When we need to obtain some kind of detector we can use the immune system. Its main function is to detect, so we can use that feature for our interest. Combining the knowledge of the immune system with the modern methods of antibody production is possible to get a large number of high-affinity recognition molecules.

To be able of all this manipulations we need to understand how bionanomachines work. To do this we can use some methods like x-ray crystallography to get atomic structures. We obtain a 3D map of the molecules that help us to understand. The resolution of this map depends of the quality of the crystals used, this is a major in this technique.

Another method is the nuclear magnetic resonance. It is used to determine molecular structure in chemistry. It characterizes the local environment of atomic nuclei inside molecules. We can alter the atomic nuclei inside molecules using a radio frequency, then, when they relax, emit radio frequency that shows the structure of the molecule. But this method is not good for big molecules, so it can be only used with some nucleic acids or small proteins. However, a 2D map can be used for greater molecules.Electron microscopy is a method that can reveal molecular morphology. It uses electrons to look at the molecules. But it has some limitations. Due to imperfections in the magnetic optics and problems with the specimen preparation it is not possible to see individual atoms, only overall morphology. The good point about this method is that it can provide information about the structure of large molecules that are impossible to study with the other previous methods. But the information can be combined with the others to get a more accuracy study.The atomic force microscopy is a method used for getting the surface of a molecule. It is like touch the molecule and trace a topographic map from that data.  This method was developed to study samples in a water environment where other methods cannot be applied. Today the problem is solved by immersing the sample in solvent. Because the samples are inside a similar environment as the one inside cells it is really good because gives data of the nanomachine as it is in its real environment.The molecular modeling using computers has revolutionized the study of biomolecules. Computer graphics allow us to visualize the molecules in a familiar manner. The best visualizations captures the keys of the molecules and shows them in 3D models that are easy to understand.Some free software available for visualize molecules is:
RasMol
ProteinExplorer
Chime
(http://www.rcsb.org/pdb/software-list.html)

The computer modeling can be used to predict biomolecular structures and functions that depends on that structures. This can be for optimization of the structure of a molecule. For normal mode analysis getting the forces working in a molecule. The molecular dynamics, how the molecule works in a variant environment, and free energy perturbation, shifting the system smoothly to learn what does the molecule.

Another commercial and academic software is:
Insight (BioSym): commercial
Sybyl (Tripos): commercial
Amber (UCSF): academic

There is a problem with proteins (the protein folding problem), we need to predict the folded structure before starting with any other study. But this is hard because every amino acid has interactions with other neighbors, so we can't know easily the structure. Another problem is to estimate the stability of the trial during the experiment. Proteins have lots of bounds inside that stabilize them. But outside water they act different.This two problems make hard to predict protein folding. Currently the best predictions are from homology modeling. Taking a similar protein and estimate the structure by similarities is easier than any other method.Simulations of biomolecular interaction are for knowing how molecules interacts with other molecules. The most successful methods combine two capabilities, a fast algorithm to search how the molecules can fit together and an energetic model that predicts the energy of the interaction. Current methods are:AutoDock (Scripps Research Institute): a genetic algorithm to evaluate energies.Dock (UCSF): a geometric matching algorithm to search in databases.Using this new functionalities new computer-assisted molecular dessign is possible. They allow us to create improved molecules. In practical application it is used for design mutations to increase stability and shifting in proteins. Many of these techiniques are used in the drug industry.

The inside and outside of a cell is water and without it, it would be impossible for them to work. Water creates a bounch of hard restrictions for the nanomachines working in atomic scale. It has strong interactions with electronic charges, zones rich in nitrogen and oxygen. This is called hidrofilic effect. However, zones rich in carbon don`t form bonds with water. That is a really important characteristic that defines most of the molecules.There are four basic molecules in every organism. Proteins, nucleic acids, polysaccharides and lipids.Most of the bionanomachines are composed of protein. They can be really rigid or not rigid at all. They can be part of structures. Enzimes are very small proteins. So there is a bounch of possibilities.The typical sice is from 200 to 500 aminoacids. This aminoacids associate themselves in chains and form variable spacial forms. The two most stable are the a-helix and the B-sheet. There are 20 different aminoacids. They combined in the proper order forms a proteine. Some of these amino acids are "special" and they can be used to stop the production. The amino acids used to start are still being studied. Other amino acids are used to another specialized tasks. There are, however, some proteins larger than 500. In some bacterias they can contain more than 2000, but when they are too big they become unstable and the errors produced during the transcription can be fatal.So protenis are really versatile, that is why bionanotechnology try to use them to construct many different molecules for many different purposes.This proteins are coded from the information stored in the nucleic acids. They are the data storage of nature. There are two types of them: DNA (desoxyribonuycleic acid) and RNA (ribonucleic acid). There are 4 bases to construct DNA, adenine, guanine, cytosine and thymine. The same are used to RNA, but instead thymine uracil is used. It is almost the same molecule, but with some changes. In each base there are 2 bits stored. There are 4 possibilities, one bit stores 2 of them, so 2 are needed. Although this nucleicacids are to store information they can be used to another purposes because of the easy predictable, strong and large structures they can form. For example, the robosoms are mostly formed by RNA.

Another molecule very common is the lipid. It is a molecule with two parts: one hydrophobic and another one hydrophilic. This feature makes them to associate in a water environment creating large molecules that can be membranes or globules. In this membranes there are inserted some other molecules that give them some special skills. Lipids are impermeable to big molecules, but permeable to the small ones.

The last molecules are the polysaccharides. This is the most versatile molecule of organisms. It can combine itself with many other molecules getting many different properties. Can be liquid like mucus or really hard like nails.

Evolution has been crucial in all this development. It is the most important and fine test to any machine. It selects the accuraced changes and kills everything not enough specialiced or useless. It provides changes that are impossible by intelligent design. Evolution places strong constraints to any change. That is why it favores modification over innovation. The nanomachines inside our body, for example, can't work in another environments outside ourselves. But because of this they have improved and have many methods to remain. They have redundancy implemented to avoid errors. In every duplication of a nanomachine, molecule or whatever errors can appear. If they appear in nucleic acids they are called mutations. This changes can be bad, meaning they can kill the new child (in an organism) or make non useful the molecule. But sometimes they can improve the function of that individual and then evolution will reward it so that change will remain in its descendants. Because all this changes are in the same enviroment the molecules working there are always in the same conditions, without them the nanomachines are not working any longer.

Nanotechnology is the ability to construct any kind of machine with a known and predefined number of atoms, instead of the traditional machines which have an arbitrary number of them. To do this, one idea is to construct an "assembler", a machine capable of making another machines. This machine was supposed to be the great new advance, with it the "2-weeks revolotion" would be possible. This is, if you have a machine able to construct another machines in a very short time you can do whatever you want. But nowadays this is far from our possibilities. That is why the bionanotechnology is important.Biology already has implemented an extraordinary number of nanomachines and has tested them during millions of years within an evolution process. This tests are much more optimized than any test we can make. In each single cell it is a really big number of nanomachines working, and they actually create another nanomachines. So nature has already created those assemblers we are looking for. Using that machines to create our own machines is bionanotechnology.Nanotechnology is a pretty new science. It began with the alchemists, then continued with the chemistry and in the 20th century it has been improved by nuclear physics.In the other hand biology began to play with molecules so the biotechnology started. If we join this two sciences we have a good way to manipulate individual atoms to create everything we can imagine.Thanks to this we can create specific nanomachines to work in many different fields, but the most important is medicine.

Felicitar a Juan, Kris y a el resto de colaboradores del blog ( y del foro ).

No dejeis de echar un vistazo a los blogs participantes. Son muchos pero es una muy buena ocasión para conocer nuevas bitácoras y siempre encuentras alguna perla. También podeis echar un vistazo a las apuestas de blogoff ( I & II ) para los premios con descripción del blog y entradas escogidas para hacerse a una idea.

[tags]Blogoff, tecnologia, 20minutos[/tags] 

Son esas pequeñas cosas negras que hay en las placas verdes que tiene cualquier cacharro electrónico dentro. Los microprocesadores de los ordenadores son lo mismo pero más "a lo bruto". Lo que voy a poner aquí es la forma en que se construían los circuitos nMOS. Siguiendo más o menos el mismo esquema se consiguen los CMOS, que fueron la evolución (bueno, estuvo EDMOS por medio) y probablemente los hayais visto en varios sitios (creo recordar que los 486 tenían de estos). Los CMOS lo mismo los explico con más detalle más adelante.

Bueno, a lo que vamos. Un circuito integrado es básicamente un montón de puertas lógicas construidas todas a la vez en un sólo chip. Lo bueno que tienen es que no hay soldaduras dentro de ellos y que el espacio entre transistores (MOSFET) es muy pequeño (los transistores forman las puertas lógicas) y que son baratos de construir (relativamente). Al no haber soldaduras (el punto más debil de todo circuito) pueden durar hasta 200 años o más, y al ser el espacio tan reducido la disipación de energía es menor (aunque aparece el problema de la refrigeración). No voy a aburriros con fórmulas, así que os lo creeis y punto ;) 

Pues bien, estos chips están hechos principalmente de silicio. Al silicio hay que "doparlo" con átomos de otros elementos para que se vuelva conductor. Es decir, lo que queremos es una base de silicio cristalino con ciertas partes dopadas, un poco de metal para ciertos contactos y óxido de silicio (que no es conductor) para aislar.

Sabiendo esto, empecemos por partes:

Cómo conseguir una base de silicio cristalino para hacer crecer(grow) el circuito encima:

Cogemos silicio (arena) y lo fundimos en un horno. Se irá separando poco a poco de los desechos que lleve y nosotros lo recuperaremos, obteniendo así silicio de grado metalúrgico (MGS). (si un día de estos estoy aburrido os pongo el esquema del horno) Teniendo ya el MGS lo metemos en un reactor de alto vacío y lo destilamos (sisi, igual que el vino en el cole). Obtenemos el EGS (silicio de grado electrónico).

Dentro del reactor lo que pasa es:

        SiHCl3 (gas) + H2 (gas) —-> Si (sol) + 3HCl (gas)

Cogemos el EGS y lo metemos en otro reactor, lo fundimos y a partir de una semilla de silicio cristalizado vamos obteniendo un cilindro de silicio cristal. Vale, llega el momento de explicar una curiosa propiedad del silicio (y cualquier cosa que cristalice). Si tienes un pequeño trozo de cristal en el que los átomos están perfectamente alineados, lo pones en contacto con silicio fundido y empiezas a enfriar lentamente los átomos irán colocándose en forma de malla siguiendo con la formación cristalina. Lo que haremos aquí será lo mismo que mojar una galleta en colacao y luego empezar a subir la galleta lentamente mientras el colacao debajo va haciendo un cilindro.

Ahora a este cilindro le pasamos unos hilos incandescentes alrededor lentamente. Se va fundiendo de nuevo pero sólo localmente y lo que harán será quedarse con toda la porquería que quede. Así tenemos ya un cilindro de perfecto cristal de silicio.

Es hora de pulir el cilindro por fuera para quitar posibles átomos rebeldes que hayan quedado en la superficie.

Y después de esto lo cortamos en pequeñas obleas finísimas, puliremos la parte superior de esta oblea y podremos comenzar a fabricar el circuito sobre ella !!! :D :D

Sobre cada una de estas obleas fabricaremos un montón de chips que luego cortaremos. Se fabrican muchos a la vez porque la mortalidad de chips es alta y sale muchísimo más barato hacer muchos a la vez.

Para la fabricación de los chips hay varios procesos que pueden utilizarse:

• Epitaxia:  se mete la oblea en un reactor. Se introduce SiH4(gas). La molécula se divide y el Si se deposita en las paredes del reactor mientras que el H2 se saca. Lo que se deposite en la oblea estará formando lo que será el chip. Hay que hacerlo despacio para que conserve la estructura cristalina. Este mismo proceso se puede hacer con SiO2 (óxido de silicio) para poner capas aislantes. Se debe hacer a la menor temperatura posible (400ºC) y muy despacio (creo que era un día lo que tardaba para unos cuantos nanómetros) para que las moléculas no se monten una fiesta y no se rompa la estructura cristalina.

• Oxidación térmica: hay dos tipos, pero son bastante malos los dos porque pueden romper la estructura cristalina (y a estas alturas ya sabremos que eso no está nada bien)

• seca: en un horno se mete la oblea junto con O2. Se forma SiO2 lentamente. Hay que tenerlo a 900-1000ºC.

• húmeda: se mete la oblea junto con agua para conseguir el mismo óxido (más H2). La temperatura es algo menor, 800-900ºC, y es más rápida que la seca, pero da peores resultados

• Impurificación:  Se trata de ponerle los dopantes al chip. Hay dos formas diferentes de hacerlo:

• Difusión: se cuece la oblea con el dopante que queramos en forma gaseosa. Poco a poco irá metiéndose en la oblea. Lo malo es que nos la podemos cargar.

• Implantación iónica: en pequeñas partes seleccionadas disparamos átomos dopantes en los sitios exactos en que los necesitemos. Aunque tenemos mucha mayor precisión así tenemos que hacerlo con el mínimo de energía posible para que los átomos que disparamos se queden en la superficie y no penetren demasiado, o incluso peor, destrocen la oblea.

• Fotolitografía: este es el bueno y me parece curiosísimo. Es la forma de poder "escribir" los circuitos en el chip y decir dónde va cada cosa. Hay que tener en cuenta que son del orden de unos cuantos nanómetros, así que no hay mucho lugar para el error.

• Primero ponemos una capa de resina por encima de toda la oblea. Esta resina es fotosensible.

• Con una máscara creada por ordenador debujamos con luz ultra violeta en la resina lo que tendrá que doparse. La máscara es una plantilla que tiene dibujos con los circuitos. Se utiliza una lente para proyectar la luz, es complicado dibujar en nanómetros ;)

• La resina en los sitios que ha sido iluminada se vuelve soluble en ácido fosfídrico, las demás zonas no. Con lo cual podemos quitar las partes que queramos de la resina

• Se quita con ácido el SiO2 que estaba debajo de la resina que acabamos de quitar

• Se implanta lo que haga falta, con dopantes, metal…lo que sea

• Se elimina el resto de la fotorresina

• se quita el Sio2 sobrante

Y después de esto varias veces en diferentes capas ya tenemos el chip. Falta cortarlos y listo. Como vereis aquí nadie mete unas pinzas, un destornillador ni nada de nada de nada. De hecho en las salas de fabricación está estrictamente prohibido que cualquier persona entre. Un pelo, un poco de piel…cualquier cosa puede ser nefasto. Se llama crecer los chips porque, es lo que se hace, ni más ni menos. Se pone una base y encima se van haciendo crecer capas de los elementos que necesitamos, hasta que tenemos el circuito integrado listo.

Por último, para poder utilizar el chip se sueldan las salidas en un sustrato (generalmente cerámico) que le proporciona sustentación mecánica y ayuda a la disipación del calor. esto es lo que vemos en la placa del ordenador. El sustrato por debajo. Si alguna vez desmontais uno y le dais la vuelta vereis que tiene un montón de pinchos. Cada uno de esos es un pin y actúa como entrada al microprocesador. El chip en sí será un pequeño cuadrado que hay en el medio de ese mazacote ;)

[tags]microelectronica, fabricacion, chips, tecnologia[/tags]

• En la Wiki : Cómo se construyen los circuitos integrados