Professional Documents
Culture Documents
Textbook Nanometer Cmos Ics From Basics To Asics 2Nd Edition Harry J M Veendrick Auth Ebook All Chapter PDF
Textbook Nanometer Cmos Ics From Basics To Asics 2Nd Edition Harry J M Veendrick Auth Ebook All Chapter PDF
https://textbookfull.com/product/bits-on-chips-harry-veendrick/
https://textbookfull.com/product/deluge-from-genesis-to-
atlantis-2019-docx-harry-sivertsen/
https://textbookfull.com/product/design-of-analog-cmos-
integrated-circuits-2nd-eddition-2nd-edition-behzad-razavi/
https://textbookfull.com/product/flash-lamp-annealing-from-
basics-to-applications-lars-rebohle/
Design of CMOS phase locked loops from circuit level to
architecture level 1st Edition Behzad Razavi
https://textbookfull.com/product/design-of-cmos-phase-locked-
loops-from-circuit-level-to-architecture-level-1st-edition-
behzad-razavi/
https://textbookfull.com/product/simpler-living-back-to-basics-
guides-2nd-edition-jeff-davidson/
https://textbookfull.com/product/covid-19-from-basics-to-
clinical-practice-1st-edition-wenhong-zhang/
https://textbookfull.com/product/lacanian-ink-29-from-an-other-
to-the-other-josefina-ayerza/
https://textbookfull.com/product/deep-learning-vol-2-from-basics-
to-practice-andrew-glassner/
Harry J.M. Veendrick
Nanometer
CMOS ICs
From Basics to ASICs
Second Edition
Nanometer CMOS ICs
Harry J.M. Veendrick
Second Edition
123
Harry J.M. Veendrick
Heeze, The Netherlands
CMOS scaling has entered the sub-20 nm era. This enables the design of system-on-
a-chip containing more than ten billion transistors. However, nanometre level device
physics also causes a plethora of new challenges that percolate all the way up to the
system level. Therefore, system-on-a-chip design is essentially teamwork requiring
a close dialogue between system designers, software engineers, chip architects,
intellectual property providers, and process and device engineers. This is hardly
possible without a common understanding of the nanometre CMOS medium, its
terminology, its future opportunities and possible pitfalls. This is what this book
provides.
It is a greatly extended and revised version of the previous edition. So besides
the excellent coverage of all basic aspects of MOS devices, circuits and systems, it
leads the reader into the novel intricacies resulting from scaling CMOS towards the
sub-10 nm level. This new edition contains updates and additional information on
the issues of increased leakage power and its mitigation, to strain induced mobility
enhancement. Immersion and double patterning litho and extreme UV and other
alternative litho approaches for sub-20 nm are extensively discussed together with
their impact on circuit layout. The design section now also extensively covers
design techniques for improved robustness, yield and manufacturing in view of
increased device variability, soft errors and decreased reliability when reaching
atomic dimensions. Both devices and ICs have entered the 3D era. This is reflected
by discussions on FinFETs, gate-all-around transistors, 3D memories and stacked
memory dies and 3D packaging to fully enable system-in-a-package solutions.
Finally, the author shares his thoughts on the challenges of further scaling when
approaching the end of the CMOS roadmap somewhere in the next decade.
This book is unique in that it covers in a very comprehensive way all aspects of
the trajectory from state-of-the-art process technology to the design and packaging
of robust and testable systems in nanometre scale CMOS. It is the reflection
of the author’s own research in this domain but also of more than 35 years of
experience in training the full CMOS chip development chain to more than 4500
semiconductor professionals at Philips, NXP, ASML, Infineon, ST Microelectron-
ics, TSMC, Applied Materials, IMEC, etc. It provides context and perspective to all
semiconductor disciplines.
v
vi Foreword
I strongly recommend this book to all engineers involved in the design, lithogra-
phy, manufacturing and testing of future systems-on-silicon as well as to engineer-
ing undergraduates who want to understand the basics that make electronics systems
work.
vii
viii Preface
2T
1 Tbit ?
1T expected capacity
256 G 256 Gbit
64 G 64 Gbit
16 G 16 Gbit
4 Gbit
4G
1 Gbit
1G
number of components per IC
Total Logic
Semicon- MOS ≈ 36 % of
Integrated (Including
ductor MOS Digital
Circuits BiCMOS)
Market ≈ 81 % of Digital Micros
≈ 98 % of the
the Total ≈ 86 % of ≈ 34 % of
integraded
Market MOS MOS Digital
Circuit
Market Memories
≈ 30 % of
MOS Digital
Fig. 3 The development of the first IC: in 1958, Jack Kilby demonstrated the feasibility of
resistors and capacitors, in addition to transistors, based on semiconductor technology. Kilby,
an employee of Texas Instruments, submitted the patent request entitled ‘Miniaturized Electronic
Circuits’ in 1959. His request was honoured. Recognition by a number of Japanese companies
in 1990 means that Texas Instruments is still benefiting from Kilby’s patent (Source: Texas
Instruments/Koning and Hartman)
are mainly digital. This reflects the fact that more than 85% of all modern CMOS ICs
are digital circuits. However, the material presented will also provide the analogue
designer with a basic understanding of the physics, manufacture and operation of
nanometre CMOS circuits. The chapters are summarised below. For educational
purposes, the first four chapters each start with a discussion on nMOS physics,
nMOS transistor operation, nMOS circuit behaviour, nMOS manufacturing process,
etc. Because the pMOS transistor operation is fully complementary to that of the
nMOS transistor, it is then easier to understand the operation and fabrication of
complementary MOS (CMOS) circuits. The subjects per chapter are chosen in
a very organised and logical sequence so as to gradually build the knowledge,
from basics to ASICs. The knowledge gathered from each chapter is required to
understand the information presented in the next chapter(s). Each chapter ends with
a reference list and exercises. The exercises summarise the important topics of the
chapter and form an important part of the complete learning process.
Chapter 1 contains detailed discussions of the basic principles and fundamental
physics of the MOS transistor. The derivation of simple current-voltage equations
for MOS devices and the explanation of their characteristics illustrate the relation-
ship between process parameters and circuit performance.
Preface xi
Fig. 5 A digital filter which comprises a few thousand transistors (Source: NXP Semiconductors)
xii Preface
Fig. 8 A 128 Gb TLC NAND flash (50 billion transistors), containing wear levelling algorithms
to increase lifetime (Courtesy of Micron Technology)
The majority of available memory types are therefore examined in Chap. 6. The
basic structures and the operating principles of various types are explained. In
addition, the relationships between their respective properties and application areas
are made clear.
Developments in IC technology now facilitate the integration of complete
system-on-a-chip, which contain several hundreds of millions to several billion
transistors. The various IC designs and realisation techniques used for these VLSI
ICs are presented in Chap. 7. The advantages and disadvantages of the techniques
and the associated CAD tools are examined. Various modern technologies are used
to realise a separate class of VLSI ICs, which are specified by applicants rather than
manufacturers. These application-specific ICs (ASICs) are examined in this chapter
as well. Motives for their use are also discussed.
As a result of the continuous increase of power consumption, the maximum level
that can be sustained by cheap plastic packages has been reached. Therefore, all
CMOS designers must have a ‘less-power attitude’. Chapter 8 presents a complete
overview of less-power and less-leakage options for CMOS technologies, as well as
for the different levels of design hierarchy.
Increased VLSI design complexities, combined with higher frequencies, create
a higher sensitivity to physical effects. These effects dominate the reliability and
signal integrity of nanometre CMOS ICs. Chapter 9 discusses these effects and the
design measures to be taken to maintain both reliability and signal integrity at a
sufficiently high level.
Preface xv
Fig. 9 Various semiconductor component sizes (e.g. atom, transistor, integrated circuit) in
perspective
Finally, testing, yield, packaging, debug and failure analysis are important factors
that contribute to the ultimate costs of an IC. Chapter 10 presents an overview of the
state-of-the-art techniques that support testing, debugging and failure analysis. It
also includes a rather detailed summary on available packaging technologies and
gives an insight into their future trends. Essential factors related to IC production
are also examined; these factors include quality and reliability.
The continuous reduction of transistor dimensions associated with successive
process generations is the subject of the final chapter (Chap. 11). This scaling has
various consequences for transistor behaviour and IC performance. The resulting
increase of physical effects and the associated effects on reliability and signal
integrity are important topics of attention. The expected consequences of and road-
blocks for further miniaturisation are described. This provides an insight into the
challenges facing the IC industry in the race towards nanometre devices.
xvi Preface
Not all data in this book completely sprouted from my mind. A lot of books
and papers contributed to make the presented material state of the art. Considerable
effort has been made to make the reference list complete and correct. I apologise for
possible imperfections.
Acknowledgements
discussions on these subjects in Chap. 10. Finally, I want to thank Chris Wyland and
John Janssen, for their remarks and additions on electrical and thermal aspects of IC
packages, respectively
I am very grateful to all those who attended the course, because their feedback
on educational aspects and their corrections and constructive criticism contributed
to the quality and completeness of this book.
In addition, I want to thank Philips Research and NXP Semiconductors, in
general, for the co-operation I was afforded. I thank my son Bram for the layout
of the cover and the layout diagrams in Chap. 4, and Ron Salfrais for the correctness
of a large part of the English text.
I would especially like to express my gratitude to my daughter Kim and Henny
Alblas for the many hours they have spent on the creation of excellent and colourful
art work, which contributes a lot to the quality and clarity of this book.
Finally, I wish to thank Harold Benten and Dré van den Elshout for their
conscientious editing and typesetting work. Their efforts to ensure high quality
should not go unnoticed by the reader.
However, the most important appreciation and gratitude must go to my family,
again, and in particular to my wife, for her years of exceptional tolerance, patience
and understanding. The year 2007 was particularly demanding. Lost hours can never
be regained, but I hope that I can give her now a lot more free time in return.
This second full-colour edition covers the same subjects, but then they are
completely revised and updated with the most recent material. It covers all subjects,
related to nanometre CMOS ICs: physics, technologies, design, testing, packaging
and failure analysis. The contents include substantially new material along with
extended discussions on existing topics, which leads to a more detailed and complete
description of all semiconductor disciplines. The result is a better self-contained
book which makes it perfectly accessible to semiconductor professionals, academic
staff and PhD and (under)graduate students.
Finally, I wish to thank Harold Benten and Kim Veendrick, again, for their
conscientious text-editing and excellent art work, respectively.
xix
xx Overview of Symbols
Dielectric constant
0 Absolute permittivity
ox Relative permittivity of oxide
r Relative permittivity
si Relative permittivity of silicon
E Electric field strength
Ec Conduction band energy level
Ef Fermi energy level
Ei Intrinsic (Fermi) energy level
Emx Maximum horizontal electric field strength
Eox Electric field across an oxide layer
Ev Valence band energy level
Ex Horizontal electric field strength
Exc Critical horizontal field strength
Ez Vertical electric field strength
Electric potential
f Fermi potential
s Surface potential of silicon w.r.t. the substrate interior
MS Contact potential between gate and substrate
F Feature size (= size of a half pitch used for stand-alone memories)
f Clock frequency
fmax Maximum clock frequency
Factor which expresses relationship between drain-source
voltage and threshold-voltage variation
gm Transconductance
I Current
Ib Substrate current
Ids Drain-source current
Ids0 Characteristic subthreshold current for gate-substrate voltage of 0 V
IdsD Driver transistor drain-source current
IdsL Load transistor drain-source current
Idssat Saturated transistor drain-source current
Idssub Subthreshold drain-source current
Imax Maximum current
Ion On current
IR Current through resistance
i.t/ Time-dependent current
j Current densisty
k Boltzmann’s constant
K K-factor; expresses relationship between
source-substrate voltage and threshold voltage
K Amplification factor
Wavelength of light
L Effective transistor channel length and inductance
Overview of Symbols xxi
Proton: a subatomic particle with a positive charge of 1:6 1019 C and a mass
of 1:67 1027 kg. Protons form, together with neutrons, the basic elements from
which all atomic nuclei are built and are held together by a strong nuclear force.
Neutron: a subatomic particle with a no charge, with a mass which is about equal to
that of a proton.
Atom: an atom is the smallest unit of any material in the periodic system of elements.
It consists of a nucleus with a fixed number of protons and neutrons, surrounded by
one or more shells, which each contain a certain number of electrons. Since an atom
is electrically neutral, the total number of electrons in these shells (one or more;
hydrogen has only one electron) is identical to the number of protons in the nucleus,
since neutrons have no net electrical charge. The number of protons in the nucleus
defines the atomic number of the element in the periodic table of elements and
determines their physical and chemical properties and behaviour. Most of the CMOS
circuits are fabricated on silicon wafers. Silicon is in group IV, which means that it
has four electrons in the outer shell. In a mono crystalline silicon substrate, each of
these four electrons can form bonding pairs with corresponding electrons from four
neighbour silicon atoms, meaning that each silicon atom is directly surrounded by
four others. By replacing some of the silicon atoms by boron or phosphorous, one
can change the conductivity of the substrate material In this way nMOS or pMOS
transistors can be created. Atom sizes are of the order of 0:1 0:4 nm. In a mono
crystalline silicon substrate the atom to atom spacing is 0:222 nm, meaning that
there are between 4 to 5 silicon atoms in one nanometer.
xxv
xxvi Explanation of Atomic-Scale Terms
Molecule: a molecule is the smallest part of a substance that still incorporates the
chemical properties of the substance. It is built from an electrically neutral group of
atoms, which are bound to each other in a fixed order. The mass of a molecule is the
sum of the masses of the individual atoms, from which it is built. A simple hydrogen
molecule (H2 ), for example, only consists of two hydrogen atoms which are bound
by one electron pair. A water molecule (H2 O) consists of two light hydrogen atoms
and one (about 16 times heavier) oxygen atom.
xxvii
Contents
1 Basic Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 1
1.1 Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 1
1.2 The Field-Effect Principle . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 1
1.3 The Inversion-Layer MOS Transistor . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 4
1.3.1 The Metal-Oxide-Semiconductor (MOS) Capacitor . . . . 9
1.3.2 The Inversion-Layer MOS Transistor .. . . . . . . . . . . . . . . . . . . 11
1.4 Derivation of Simple MOS Formulae . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 18
1.5 The Back-Bias Effect (Back-Gate Effect, Body Effect)
and the Effect of Forward-Bias . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 21
1.6 Factors Which Characterise the Behaviour of the MOS
Transistor . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 25
1.7 Different Types of MOS Transistors . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 26
1.8 Parasitic MOS Transistors . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 28
1.9 MOS Transistor Symbols . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 29
1.10 Capacitances in MOS Structures . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 30
1.11 Conclusions .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 38
1.12 Exercises .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 39
References .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 43
2 Geometrical-, Physical- and Field-Scaling Impact on MOS
Transistor Behaviour .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 45
2.1 Introduction .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 45
2.2 The Zero Field Mobility . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 45
2.3 Carrier Mobility Reduction . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 46
2.3.1 Vertical and Lateral Field Carrier Mobility Reduction . . 47
2.3.2 Stress-Induced Carrier Mobility Effects . . . . . . . . . . . . . . . . . 50
2.4 Channel Length Modulation . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 51
2.5 Short- and Narrow-Channel Effects . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 53
2.5.1 Short-Channel Effects .. . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 53
2.5.2 Narrow-Channel Effect . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 55
2.6 Temperature Influence on Carrier Mobility and Threshold
Voltage .. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. . . . . . . . . . . . . . . . . . . 57
xxix
xxx Contents
4. The “Paradiso”
Structure.—Dante’s Paradise consists of the nine moving
heavens, according to Ptolemaic astronomy, crowned by the tenth
motionless and divinest Empyrean heaven, “according to what Holy
Church teacheth, who cannot lie” (Conv. ii. 3, 4). The nine moving
spheres revolve round our globe, the fixed centre of the Universe,
each of the lower eight being enclosed in the sphere above itself.
The seven lowest are the heavens of the planets: the Moon,
Mercury, Venus, the Sun, Mars, Jupiter, Saturn. The eighth or stellar
heaven, the sphere of the Fixed Stars or Firmament, is the highest
visible region of the celestial world, and to some extent corresponds
to the Earthly Paradise in the lower realms. Above this visible
firmament, the ninth or Crystalline heaven, the Primum Mobile,
directs with its movements the daily revolution of all the others. In it
nature starts; from it proceed time and motion, with all celestial
influence for the government of the world (Par. xxvii. 106-120). It is
“the royal mantle of all the volumes of the world, which is most
fervent and most living in God’s breath, and in His ways” (Par. xxiii.
112-114); and it communicates in different degrees some
participation in this quickening breath of God to the other sphere
which it encloses, and to all the Universe. It moves swiftest of all,
from the fervent desire of all its parts to be united to the Empyrean,
the spaceless and motionless ocean of Divine love, where God
beatifies the saints and Angels in the vision of His Essence. This
Empyrean is the true intellectual Paradise, for which the lower
heavens are merely sensible preparations. “This is the sovereign
edifice of the world, in which all the world is included, and outside of
which is nothing; and it is not in space, but was formed only in the
First Mind” (Conv. ii. 4); “The heaven that is pure light; light
intellectual full of love, love of true good full of joy, joy that
transcendeth every sweetness” (Par. xxx. 39-42).
Gradations.—Each of the nine lower spheres represents a step
higher in knowledge, in love, in blessedness, until in the true
Paradise the soul attains to perfect knowledge, supreme love, and
infinite blessedness in union with the First Cause, in the Beatific
Vision of the Divine Essence. The ascent is marked by the increased
loveliness of Beatrice, as she guides Dante upwards from heaven to
heaven; it is marked, too, by gradations in the brilliancy of the
blessed spirits themselves, by their ever increasing ardour of charity
towards the poet, and by the growing spirituality of the matters
discussed in each sphere—veil after veil being drawn aside from the
mysteries of the Divine treasure-house.
The Saints.—“To show forth the glory of beatitude in those souls,”
says the letter to Can Grande, “from them, as from those who see all
truth, many things will be sought which have great utility and delight”
(Epist. x. 33). All the saints without exception have their home and
glorious seats with Mary and the Angels in that Empyrean heaven,
where they are finally seen as glorified spirit likenesses of what they
were on earth. But into each preparatory sphere, excepting the ninth,
these citizens of eternal life descend to meet Dante as, with
Beatrice, he approaches the gates of the celestial city—like the
noble soul returning home to God in the fourth and last part of life:
“And even as its citizens come forth to meet him who returns from
a long journey, before he enters the gate of the city, so to the noble
soul come forth, as is fitting, those citizens of eternal life. And thus
they do because of her good works and contemplations; for, being
now rendered to God and abstracted from worldly things and
thoughts, she seems to see those whom she believes to be with
God” (Conv. iv. 28).
In all these spheres, excepting the first, and to some extent the
second, the spirits of the blessed appear clothed in dazzling light,
which hides their proper semblances from Dante’s gaze, making
them appear as brilliant stars or flaming splendours. In the tenth
Heaven of Heavens he is supernaturally illumined, and enabled
thereby to behold them in their glorified spirit forms “with
countenance unveiled” (Par. xxii. 60, xxx. 96, xxxi. 49).
In the three lower heavens, to which earth’s shadow was
supposed to extend (Par. ix. 118, 119), appear the souls whose lives
were marred by inconstancy in their vows, who were moved by vain
glory, or yielded to sensual love. They descend into these lower
spheres to give Dante a sensible sign of the lesser degree of the
perfection of their beatitude in the Empyrean. Domus est una, sed
diversitas est ibi mansionum; “The house is one, but there is a
diversity of mansions there.” There are different mansions of
beatitude in God’s house, proceeding from inequality in the soul’s
capacity of the Divine Charity; but in that house all are fulfilled with
the Vision of the Divine Essence, and each perfectly beatified
according to his own capacity of love and knowledge. In the spheres
of the four higher planets appear the souls of great teachers and
doctors, of Jewish warriors and Christian knights, of just rulers, of
ascetic monks and hermits; they appear as types of lives perfected
in action or in contemplation, as a sign of the different ways in which
perfection may be reached on earth and beatitude attained in
Paradise. These successive manifestations in the seven spheres of
the planets obviate what might otherwise have proved the monotony
of a single heaven, and suggest that, although each soul partakes
supremely according to its individual capacity of the Beatific Vision,
which is essentially one and the same in all, yet there are not only
grades but subtle differences in the possession of it, in which the life
on earth was a factor. In the eighth, the Stellar Heaven, still under
sensible figures and allegorical veils, Dante sees “the host of the
triumph of Christ, and all the fruit gathered by the circling of these
spheres” (Par. xxiii. 19-21), representing the Church in which these
various modes and degrees of life are brought into unison. In the
ninth, the Crystalline, the angelic hierarchies are manifested with
imagery symbolical of their office towards God and man,
representing the principle of Divine Order, the overruling and
disposition of Divine Providence in which the celestial intelligences
are the agents and instruments. The Empyrean Heaven depicts the
soul in patria, with all the capacities of love and knowledge
actualised in the fruition of the Ultimate Reality, the supreme and
universal truth which is the object of the understanding, the supreme
and universal good which is the object of the will.
The Angels.—Each of the nine moving spheres is assigned to the
care of one of the nine angelic orders: Angels, Archangels,
Principalities; Powers, Virtues, Dominations; Thrones, Cherubim,
Seraphim. And the character of the blessed spirits that appear to
Dante in each heaven, and the subjects discussed, seem in almost
every case to correspond more or less closely with the functions
assigned by mystical theologians, especially Dionysius, St. Gregory
and St. Bernard, to the special angelic order which presides over the
sphere in question. There are two fundamental principles in the life
of the soul: nature and grace. The one is represented in the Paradiso
by the astronomical order of the heavens and their influence upon
individual disposition, furnishing man with a natural aptitude for the
moral and intellectual virtues; the other by the bounty of Divine
Grace, which reveals itself in the perfecting of the natural and the
infusion of the supernatural virtues, whereby souls become
assimilated to the angelic orders.[37] It is through these Angels (the
name is applied generally to all, as well as to the lowest order) that
God disposes the visible world; in the hands of the celestial
intelligences the heavens are as hammers, to stamp the Divine ideas
upon material creation and carry out the Divine plan in the
government of the Universe (cf. Par. ii. 127-129). And, by means of
the influence of the stars, these Angels have impressed certain men
with their own characteristics; perhaps to fill up the vacant places in
their ranks left by the fall of Lucifer’s followers, certainly to co-
operate on earth in their work. Dante himself was born beneath the
constellation of the Gemini, the glorious stars impregnated with the
virtue of the Cherubim who rule the eighth sphere (Par. xxii. 112-
123). The Cherubim represent the Divine Wisdom; their name
signifies plenitude of knowledge. According to St. Bernard, they
“draw from the very fountain of wisdom, the mouth of the Most High,
and pour out the streams of knowledge upon all His citizens.” Their
special prerogatives are fullness of Divine light, and contemplation of
the beauty of the Divine order of things; they see most into the
profound mysteries of the hidden things of God, and spread the
knowledge of Him upon all beneath them. By their inspiration Dante
co-operated in this cherubical work by writing the Divina Commedia.
The Seraphim especially represent the Divine Love. No soul appears
in the ninth heaven which they guide and in which the angelic
hierarchies are manifested; Beatrice is the sole interpreter between
the poet and the Angels, as she had been the revealer to him on
earth of Love’s “possible divinities and celestial prophecies.”
Time in Paradise.—The action of the Paradiso begins at noon,
immediately after Dante’s return from Eunoë; that is, noon on
Wednesday in Easter week in the Earthly Paradise and (the
following) midnight at Jerusalem (Par. i. 37-45). The time-references
in this third Cantica are rather doubtful (Par. xxii. 151-153, xxvii. 77-
87), but it seems probable that Dante takes twenty-four hours to
ascend through the nine material heavens to the Empyrean, which is
beyond time and space, where “the natural law in nought is relevant”
(Par. xxx. 123). When Dante woke from his “mighty trance” to the
“sound of the importunate earth,” it was perhaps about dawn on the
morning of Friday in Easter week in our world, thus completing the
seven days of his ecstatic pilgrimage, which had begun at about the
same hour on Good Friday.
Canto I.—In a lyrical prologue of stately music (Par. i. 1-36), the
poet sings of the glory of the First Mover, and prays for light and
inspiration to complete this third most arduous portion of his divine
poem. Then, in the noblest season of the year and noblest hour of
the day, as Beatrice gazes upon the sun and Dante upon her, his
mind becomes godlike, and he ascends to Heaven swifter than
lightning. To explain his ascent, Beatrice discourses upon the form
and order of God’s visible image, the Universe; and on His Eternal
Law, the sovereign plan of government existing in the Divine Mind, to
which all movements and actions of nature are subject (ibid. 103-
141). To all created things God has given an instinct, or principle of
inclination, by which, in different ways according to their nature, He
draws them all back to Himself over the great sea of being. Rational
beings alone can resist the order of the Universe and defeat the
Eternal Law by sin, which is expiated by temporary or eternal
suffering, as Dante has seen in the lower realms; but the purified
soul, in accordance with this order and law, inevitably mounts up to
find its rest in union with the First Cause. It is the doctrine of spiritual
gravitation (derived from St. Augustine), according to which the soul
is moved by love as bodies are by their weight, and all things find
their rest in order.
The Heaven of the Moon.—They are received into the eternal
pearl of the Moon (Par. ii.); where Beatrice first confutes Dante’s
former theory concerning the luminous substance of the celestial
bodies, and, by explaining how everything in the visible world
depends upon the angelic movers of the sphere, gives a mystical
interpretation of a natural phenomenon, on this first step of his
ascent to the suprasensible. Within this eternal pearl appear faint but
divinely beautiful forms of women; the souls of those who had
yielded to violence and broken their solemn vow (Par. iii.). Piccarda
Donati, sister of Corso and Forese, sets forth the perfection of
celestial charity, where all wills are made absolutely one with the will
of God, who has awarded different degrees or mansions of beatitude
to all His chosen ones:
“And His will is our peace.”[38] Transfigured now with ineffable joy,
Piccarda tells the pathetic story of her frustrated life on earth; and
points out to Dante the Empress Constance, mother of Frederick II.,
torn, like her, from the convent’s shelter. Beatrice explains to the
poet the place of all the saints in the Empyrean—the “heaven of
humility where Mary is,” as Dante had sung long before of Beatrice
herself in the Vita Nuova—and the reason of this temporary
apparition in the moon (Par. iv.). The other questions solved in this
sphere are all connected with Free Will. Rectitude of will is
necessary for the gaining of Paradise, and nothing whatever can
take away that freedom of the will. “As regards the proper act of the
will, no violence can be done to the will”; and, since Piccarda and
Constance yielded through fear of greater evil, they fell voluntarily
from the state of perfection to which they were called. Freedom of
the will is God’s greatest gift to man (Par. v. 19-24); hence the
sanctity of an accepted vow, wherein this supreme gift is offered to
God as victim, although Holy Church has power to commute, save,